journal of technical analysis (jota). issue 56 (2001, winter)

SM

A Publication of

2001

MARKET TECHNICIANS ASSOCIATION, INC.One World Trade Center ● Suite 4447 ● New York, NY 10048 ● 212/912-0995 ● Fax: 212/912-1064 ● e-mail: [email protected] ● www.mta.org

A Not-For-Profit Professional Organization ● Incorporated 1973

MTAJOURNAL

MTA JOURNAL • Fall-Winter 2001 1

THE MTA JOURNAL – TABLE OF CONTENTSFALL-WINTER 2001 • ISSUE 56

12

34

5

THE MTA JOURNAL EDITORS AND REVIEWERS 3

MTA MEMBER AND AFFILIATE INFORMATION 4

MTA 2001-2002 BOARD OF DIRECTORS AND MANAGEMENT COMMITTEE 5

EDITOR’S COMMENTARY 6

Henry O. Pruden

EDITORS’ FAREWELL 6

Henry (Hank) O. Pruden, Ph.D., Editor & David L. Upshaw, CFA, CMT, Associate Editor

THE ART OF TECHNICAL ANALYSIS 7

Harry W. Laubscher

TECHNICAL VERSUS FUNDAMENTAL ANALYSIS: A VIEW FROM ACADEME 9

Hamid B. Shomali, Ph.D.

A DOW THEORY UPDATE 11

Ralph J. Acampora and Rosemarie I. Pavlick

IT PAYS TO BE CONTRARY 15

James L. Fraser

TEN WAYS TO PROFIT FROM A “RUNAWAY BULL MARKET” 21

James Dines

ANATOMY OF A TRADING RANGE 23

Jim Forte6

2 MTA JOURNAL • Fall-Winter 2001

8

9

10

7ANSWERING THE BELL OF SENTIMENT INDICATORS 31

Brent L. Leonard

COMBINING TECHNICAL ANALYSIS WITH FUNDAMENTAL VALUATION TO CREATE ARISK INDICATOR FOR THE STOCK MARKET 39

Jurrien H. Timmer, CMT

THE STORY OF THE THREE STOCK MARKET BOTTOMS:THE PAPA BOTTOM, THE MAMA BOTTOM AND THE BABY BOTTOM 47

Kenneth Safian

EXPLOITING VOLATILITY TO ACHIEVE A TRADING EDGE USING AN AVERAGE-TRUE RANGE (ATR)SECOND FILTER: MARKET-NEUTRAL/DELTA-NEUTRAL TRADING USING THE PRISM TRADING SYSTEMS 55

Jeff Morton, M.D., CMT and Randi Schea, M.D.


EDITOR

Henry O. Pruden, Ph.D.Golden Gate University

San Francisco, California

ASSOCIATE EDITORS

David L. Upshaw, CFA, CMT Jeffrey Morton, M.D. CMTLake Quivira, Kansas PRISM Trading Advisors

Missouri City, Texas

Connie Brown, CMTAerodynamic Investments Inc.

Pawley's Island, South Carolina

John A. Carder, CMTTopline Investment Graphics

Boulder, Colorado

Ann F. Cody, CFAHilliard Lyons

Louisville, Kentucky

Cynthia KaseKase and CompanyAlbuquerque, NM

Charles D. Kirkpatrick, II, CMTKirkpatrick and Company, Inc.

Chatham, Massachusetts

Cornelius LucaBridge Information Systems

New York, New York

Theodore E. Loud, CMTTel Advisor Inc. of Virginia

Charlottesville, Virginia

John McGinley, CMTTechnical Trends

Wilton, Connecticut

Michael J. Moody, CMTDorsey, Wright & Associates

Pasadena, California

Richard C. Orr, Ph.D.ROME Partners

Marblehead, Massachusetts

Robert B. PeirceCookson, Peirce & Co., Inc.

Pittsburgh, Pennsylvania

Kenneth G. Tower, CMTUST Securities

Princeton, New Jersey

J. Adrian Trezise, M. App. Sc. (II)Consultant to J.P. Morgan

London, England

PRODUCTION COORDINATOR

Barbara I. GompertsFinancial & Investment Graphic Design

Marblehead, Massachusetts

MANUSCRIPT REVIEWERS

THE MTA JOURNALFALL-WINTER 2001 • ISSUE 56

The Market Technicians Association Journal is published by the Market Technicians Association, Inc., (MTA) 74 Main Street,3rd Floor, Woodbridge, NJ 07095. Its purpose is to promote the investigation and analysis of the price and volume activities ofthe world's financial markets. The MTA Journal is distributed to individuals (both academic and practitioner) and libraries in

the United States, Canada, Europe and several other countries. The MTA Journal is copyrighted by the Market TechniciansAssociation and registered with the Library of Congress. All rights are reserved.

PUBLISHER

Market Technicians Association, Inc.74 Main Street, 3rd FloorWoodbridge, NJ 07095


Members Affiliates■ Invitation to MTA educational meetings ■■ ■■

■ Receive monthly MTA newsletter ■■ ■■

■ Receive MTA Journal ■■ ■■

■ Use of MTA library ■■ ■■

■ Participate on various committees ■■ ■■

■ Colleague of IFTA ■■ ■■

■ Eligible to chair a committee ■■

■ Eligible to vote ■■

Annual subscription to the MTA Journal for nonmembers: $50 (minimum two issues).Single issue of the MTA Journal (including back issues): $20 each for members and affiliates and

$30 for nonmembers.

✔ ✔

✔ ✔

✔ ✔

✔ ✔

✔ ✔

✔ ✔

✔ ■■

✔ ■■

MARKET TECHNICIANS ASSOCIATION, INC.MEMBER AND AFFILIATE INFORMATION

MEMBER

Member category is available to those "whose professionalefforts are spent practicing financial technical analysis thatis either made available to the investing public or becomesa primary input into an active portfolio management pro-cess or for whom technical analysis is a primary basis oftheir investment decision-making process." Applicants forMember must be engaged in the above capacity for fiveyears and must be sponsored by three MTA Members fa-miliar with the applicant's work.

AFFILIATE

Affiliate status is available to individuals who are interestedin technical analysis, but who do not fully meet the re-quirements for Member, as stated above; or who currentlydo not know three MTA members for sponsorship. Privi-leges are noted below.

DUES

Dues for Members and Affiliates are $200 per year and arepayable when joining the MTA and thereafter upon re-ceipt of annual dues notice mailed on July 1. College stu-dents may join at a reduced rate of $50 with the endorse-ment of a professor.

APPLICATION FEES

Applicants for Member will be charged a one time,nonrefundable application fee of $25; no fee for Affiliates.

BENEFITS OF THE MTA

STYLE OF THE JOURNAL'S AUTHORS

You want your article to be published. The staff of theMTA Journal wants to help you. Our common goal can beachieved efficiently if you will observe the following con-ventions. You'll also earn the thanks of our reviewers, edi-tors, and production people.1. Send your article on a disk or via email (but, a hard

copy is still necessary). Footnotes and references shouldappear at the end of your article.

2. Submit two copies of your article.3. All charts should be provided in camera-ready form (or

in an eps, jpg, etc.) and be properly labeled for textreference. Try to avoid using "above" or "below," butrather, Chart A, Table II, etc.

4. Greek characters should be avoided in the text and inall formulae.

5. Include a short (one paragraph) biography. We willplace this at the end of your article. Your name willappear beneath the title of your article, be sure to in-clude CMT, CFA, Ph.D., etc.We will consider any article you send us, regardless of

style, but upon acceptance, we will ask you to make yourarticle conform to the above conventions.

Send your non-CMT manuscripts tothe MTA Journal editor:

Charles D. Kirkpatrick II, CMTKirkpatrick & Company, Inc.

P.O. Box 699, Chatham, MA 02633-0699


Director: PresidentRalph J. Acampora, CMTPrudential Securities Inc.

212/778-2273, Fax: 212/778-1208E-mail: [email protected]

Director: Vice PresidentRichard A. Dickson

Hilliard Lyons502/588-4122, Fax: 502/588-9132

E-mail: [email protected]: Treasurer

Andrew BekoffVan Der Moolen LLC


Director: SecretaryKeenan Hauke

Samex Capital Partners317/566-2162, Fax: 317/816-7001E-mail: [email protected]

Director: Past PresidentPhilip B. Erlanger, CMT

Phil Erlanger Research Co., Inc.978/263-2536, Fax: 978/266-1104

E-mail: [email protected]:

Mike EpsteinNDB Capital Markets Corporation617/753-9910, Fax: 617/753-9914E-mail: [email protected]

Bruce Kamich, CMTReuters America Inc.


Philip J. Roth, CMTMorgan Stanley


Kenneth G. Tower, CMTUS T Securities Corp.


MARKET TECHNICIANS ASSOCIATION, INC.2001-2002 BOARD OF DIRECTORS & MANAGEMENT COMMITTEE

Management Committee(4 Officers, Past President and Committee Chairs)

AccreditationJ. Les Williams, CMT

Williams Capital Management, Inc.817/548-8332, Fax: 817/548-9289

E-mail: [email protected]

Fred G. Schutzman, CMT212/832-6268, Fax: 212/832-6288

E-mail: [email protected] of Knowledge

John C. Brooks, CMTYelton Fiscal Inc.


Distance LearningRichard A. Dickson

Hilliard Lyons502/588-4122, Fax: 502/588-9132


Philip J. Roth, CMTMorgan Stanley


Ethics & StandardsNeal Genda

National City Bank310/888-6416, Fax: 310/888-6388


Bruce Kamich, CMTReuters America Inc.


IFTA LiaisonHenry (Hank) O. PrudenGolden Gate University


InternshipJohn Kosar, CMT

Arbor Research & Trading847/304-1550, Fax: 847/304-1595


Charles D. Kirkpatrick II, CMTKirkpatrick & Company, Inc.

508/430-8668, Fax: [email protected]

Board of Directors(4 Officers, 4 Directors & Past President)

LibraryDaniel L. Chesler, CTA, CMT


MarketingMichael N. Kahn

516/692-2435E-mail: [email protected]

MembershipLarry Katz

Market Summary & Forecast805/370-1919, Fax: 509/693-7473


Anthony F. DwyerKirlin Holdings


PlacementRick BensignorMorgan Stanley


Programs (NY)Bernard Prebor201/434-6224


M. Frederick Meissner404/875-3733


Charles S. Comer, CMTideaglobal.com

212/271-0769, Fax: 212/571-4334Email: [email protected]

SeminarHerbert G. Labbie, CMT

Technical Portfolio Strategies412/391-3560, Fax: 412/391-2242


Philip B. Erlanger, CMTPhil Erlanger Research Co., Inc.



EDITORS’ FAREWELL

Henry (Hank) O. Pruden, Ph.D., Editor & David L. Upshaw, CFA, CMT, Associate EditorOur deep and sincere thanks to the referees and staff of the MTA Journal, most notably Barbara Gomperts.We are pleased to entrust the Editorship of the MTA Journal to the capable hands of Charles D. KirkpatrickII, CMT. And with the following, slightly modified verse from Rudyard Kipling, we bid our adieu.

The Long TrailThere's a whisper down the field where the year has shot her yield,

And the ricks stand grey to the sun,Singing: "Over then, come over, for the bee has quit the clover,

"And your eleven years are done."

You have heard the beat of the offshore wind,And the thresh of the deep-sea rain;You have heard the song - how long? how long?Pull out on the trail again!

Ha' done with the Tents of Shem, dear lass,We've seen the seasons through,And it's time to turn on the old trail, our own trail that is always new!

It's North you may run to the rime-ringed sunOr South to the blind Horn's hate;

Or East all the way into Mississippi Bay,Or West to the Golden Gate -

Were the blindest bluffs hold good, dear lass,And the wildest tales are true,

And the men bulk big on the old trail, our own trail, the out trail,And life runs large on the Long Trail - the trail that is always new.

EDITOR’S COMMENTARY

Henry O. Pruden

A DYNAMIC BODY OF KNOWLEDGEThe six articles, reprinted in this issue, span the life of the existence of the MTA Journal. These articles givetestimony to the growing, dynamic quality of the body of knowledge we know as Technical Analysis. Thefirst two articles reflect the dynamic balance between the artful practitioner (Laubscher) and the scientificacademic (Shomali); Ralph Acampora's classic on Dow Theory appeared in the very first issue of theJournal. This broad treatise on the market (joined by Dines) is followed by more specialized studies such asFraser and Leonard on sentiment, Forte on pattern recognition and Timmer on combining technicalindicators and fundamental information. And these articles are just a sample of the publications from theMTA Journal. Some articles were selected from a list assembled by the MTA Educational Foundation;others were chosen by the Editor. The two research notes by Safian and Morton are fresh contributions.


THE ART OF TECHNICAL ANALYSIS

Harry W. Laubscher 1THIS ARTICLE APPEARED IN THE MTA JOURNAL, FEBRUARY 1986

Having spent almost twenty-nine of the last thirty years working inthe stock market’s highways and byways, I hope that I have learnedsomething. We all manage to learn a great deal, regardless of whatfield we work in, but all too often many of us tend to forget many ofthe things that were learned and which should not have been forgot-ten. It has been often said that the stock market, along with drinkand women, is one of the great levelers of our time. Many of theimportant pieces of market lore that we learned along the way, andthen in later years tended to forget, no doubt could have saved manyof us from experiencing many of the mistakes that all of us make. Iam reminded of this lately as I see a great rush on the part of inexpe-rienced brokers and traders to be “in the crowd” regardless of wherethat crowd is headed. For some strange reason, the more the stockmarket rises, the more bullish many of us tend to become, finallyresulting in a great rush to own shares right at the top of the market.On the other hand, it usually works out that the lower the marketgoes in bear markets, the more bearish more people tend to become.It has always been so, and as long as people are the driving forcebehind all market movements, up or down, it will always be so.

We have all heard some of the sayings for which Wall Street is sowell known, such as “sell on the good news and buy on the bad news.”I’ve found that this does, indeed, work out to one’s benefit moreoften than not. During the Union Carbide fiasco in India when theshares of the company dropped sharply to near 33, the wise peoplewere there buying all they could get, knowing full well that the lem-ming instinct had once again taken things too far. The recovery inprice of those shares since then is in the record for all to read andgreat profits have been made in what “everyone knew” was going tobe a disaster for the company. More recently, we have the situationof Texaco and Pennzoil. Many savvy traders recognized the rathersilly awarding to Pennzoil of several billions of dollars as an opportu-nity to acquire an historically “good” company at what appeared tobe bargain priced levels. As of this writing, the shares of Texaco arestill floundering near the 30-31 level, and although my point andfigure work suggests a potential downside count to approximatelythe 29 level, I am advising investors with some patience to start ac-quiring Texaco shares in the 30-31 area. In time, this should workout to be a good buy. I use it as another example of the unsophisti-cated atmosphere that appears to be so prevalent today.

And yet, I stop to wonder if I ever really did meet anyone at allwho could accurately be described as sophisticated in the stock mar-ket. Being sophisticated in the stock market probably is as out ofplace as being logical. And we all know that in order to be successfulin the world of investing, logic has to be left outside the door. Thisbrings me around to the inevitable question: “Is understanding thestock market now becoming more of a science and less of an art?”No doubt, it is a question that has troubled the minds of manymarketeers for many years. I know, as a result of my recent trip toJapan, that the Japanese believe that scientific applications can beapplied to the stock market, and they have gone to great lengths inemploying those applications. More than any group I know, the Japa-nese are attempting to make it more of a science than it has been.And yet, I know that whenever you have to deal with something thatinvolves people to any great extent, science can only be carried so farbefore art has to take over. Thanks to many of the new inventions

that have come along over the years, such as the price quoting ma-chines, information is much more readily available, making the for-merly onerous job of keeping up to date much less so. Today, a greatdeal of information is quickly available – perhaps too much so – andthus the odds in decision-making have increased on the side of error.Now I know that that last sentence doesn’t seem right somehow, butthen you are still thinking logically, aren't you? And that doesn’twork in regard to the market. Too much information, too easily ob-tained leads one into too many possible byways, and therefore, in-creases the chances for error. Too many people believe that the moreyou know about something, the better off you are apt to be. I thor-oughly agree, except in the stock market. In this arena, one oftencan lose sight of the forest for all the trees that are available and itoften helps to use less data and a bit more gut feel. And, very often,who you know is just as important, if not more so, than what youknow. How else do you explain the success levels of those who tend tomake it in the market?

And this brings me to one of my favorite sayings about the mar-ket. It is one in which I thoroughly believe and have seen the work-ings of it spread far and wide, among all types of marketeers. “Thestock market is one of the easiest places in the world to get rich.” Allyou have to do to make it so is to avoid what most of the others aredoing. For example, I long ago gave up buying The Wall Street Journal.It has too much information of too little worth and not enough ofthe really valuable stuff. Barron's is somewhat better in that respect,but the new newspaper Investors Daily has it all over both of the DowJones papers. Once you start getting really good news on what isgoing on in the market, the path to wealth is soon beneath your feet.It helps also to look around you, ask the man in the street what hethinks about the economy, or whatever, and when you have deter-mined what the general drift of conventional wisdom is, go the oppo-site way. One of the biggest obstacles to obtaining wealth in the stockand bond markets is to fall prey to the enticements of quick profits.Of course, they are grand to have but more often than not it pays to“let your profits run, while making all endeavors to cut losses short.”Too often, technically oriented traders and investors see more in thechart than really is there to be seen. False breakouts, up or down,make us nervous and we jump, only to find out later on that therewas no alarm except in our own minds.

I also believe that it is a bit wise to be skeptical of almost every-thing. At times you will have to depend on what appears to be thewisdom of those around you, those in whom you have faith to do thejobs with which they are involved. But one also should take the timeto hear what others have to say, and then go and do a bit of “checkingit out.” It can't hurt. I also believe that too many investors fall prey tothe belief that information on revenues, management, contracts, in-dustry items, sales and earnings are what makes stocks move up anddown. They seldom stop to think that all that kind of informationonly has to do with the company itself, NOT the shares of the com-pany in question. The only thing that makes shares move up or downin price is buying or selling pressure outweighing one another. Ifnobody sells, then all the good information on dividends and earn-ings isn’t going to move shares upward. If all the bad news makesbuyers disappear, then shares can't move downward. So trying togauge what the buying pressure is probably is the most important


thing that anyone can do in the search for profitability. AtPaineWebber, every week we publish a relative strength analysis ofover 4000 issues that when taken in conjunction with some other in-formation, affords a very good indication of whether or not buying,or selling, pressure is rising or declining. Once you have that tool inyour hands, the game becomes a lot easier. Over the last ten-elevenyears, every single issue recommended in my Trends & Opportuni-ties Market has been based on my reading of the buying or sellingpressures. That has helped us achieve a 95% success ratio in thoserecommendations, 250 profits, nine losses and three unchanged since1974, regardless of whether a bull market or a bear market was in thedriver’s seat. Get to know what direction the pressure is moving inand you are halfway to your objectives. And that goes just as impor-tantly for short-selling.

While we are on the subject of short selling, technical analysis canbe of great help in helping clients make money on the short side.Once you find a chart pattern that is descriptive of distribution, moveon to find out if the selling pressure has been increasing, or if thebuying pressures are ebbing. If both suggest you should be shortingthe stock, go a step further and check out the short interest. If it ishigh, so much the better, since most of the shorting is still done byprofessionals. And don’t fall for that old saw about stocks with highshort interest holding up well, because there is a buyers’ floor underthe price. It is quite true that those who sell short must sooner orlater buy back in again in order to take their profit, or their loss. Buta check of past bear markets will show that stocks that had the high-est levels of short-interest usually sold off quite nicely, enabling thosewho sold short to repurchase shares at lower levels. A floor understocks with high short interest is about as fleeting as support levels ina bear market. I always try and remind brokers who ask me aboutsupport levels in a bear market that support is only a seven-letterword and usually doesn’t afford the support sought. Support, on theother hand, is much more important, technically, in a rising market.The same goes for resistance levels. In bull markets, those resistancelevels usually provide only fleeting roadblocks to advances. In bearmarkets, upside resistance takes on much more power, on average.There always are exceptions, of course.

I guess if I had only one tool to select from all those that are avail-able among the various charts and chart services, I would come downon the side of a good weekly bar line service. Something like Mansfieldthat provides the relative strength indicator graphically presented,

various moving averages, and then throws in upside volume and down-side volume to make it a bit easier. If you like having the fundamen-tals, those are provided as well. They once used to give earnings’estimates, but not anymore. Too bad! It helped to gauge thingsbetter if you knew what the “street” was expecting. Then, when thewinds of winter were blowing and I was all snug by my fireside, I’dtake out my barline book of charts and go through it every week,looking for those seven cardinal patterns that indicate either accu-mulation or distribution. You all know what they are. You don’tneed me taking up valuable space to repeat them again. Once I wasable to correctly identify some of those patterns, I would put some ofmy funds to work. I guess that when push comes to shove, thoseimportant patterns of accumulation or distribution are the most im-portant things in our world of Technical Analysis. Without the knowl-edge of them, we’re always back at square one.

Now I certainly don’t mean to knock point and figure analysis. Ihave found it to be too helpful over the years to give it a positionbelow the salt. It is an invaluable tool in trying to gauge just how fara move is going to carry once that move has started. But, if pressed,I still would have to say that a bar line chart will tell you when themove is going to start. Then you would move on to a P&F chart. Ifanyone out there wants to make a lot of money in this business, Iwould suggest that they start a point and figure weekly chart serviceof the 500 most actively traded listed bonds. As far as I know, there isno such service available. Since we are in the still early stages of asuper cycle bull market in bonds, their price performance will be-come increasingly important as the next five to seven years roll by.And their volatility also will increase, making a point and figure analysisfar more valuable. I’ve thought of doing it myself, but am just tootired to take on another chore.

The twenty-nine years will soon be rolling into a nice round thirty.I came to Wall Street intending to stay only twenty years, but the workwas so interesting and the people with whom I worked were so pleas-ant and helpful, that I stayed on and on and on. This has been themost fascinating of my three careers and I am wondering if the fourthand next career will be as rewarding.

BIOGRAPHY

Harry W. Laubscher is a member of the MTA and a Techni-cal Analyst at Tucker Anthony, New York, NY.


In the last several decades, there have been two competing schoolsof thought regarding the analysis and valuation of financial securi-ties. The traditional finance experts have espoused fundamental fi-nancial analysis, dealing with identification of variables that will de-termine the underlying value of securities. These traditional securityanalysts have downplayed the significance and the relevance of theother school, namely “technical analysis.” The strict fundamentalistshave viewed technical analysts as mere “chartists” who pass over pastdata in order to find certain patterns in the behavior of security pricesover time. These traditional finance theorists at times have likenedthe technical analysts to “astrologers” in the field of finance. Thetechnical analysts, on the other hand, would like to gain recognitionfor their successes in forecasting security prices and be consideredmore like astronomers than astrologers. But the debate continues.

In recent years technical analysis has been gaining wider accep-tance in academia. Technical analysis received prominent and favor-able review in a seminal article surveying the frontiers of finance whichappeared in the October 21,1993 issue of The Economist. The offer-ing of courses in technical analysis at some universities such asDartmouth College, Golden Gate University and the McIntire Schoolat the University of Virginia, as well as the New York Institute of Fi-nance, demonstrates the increasing recognition of the field of tech-nical analysis by academia. As technical analysts align their field with“behavioral finance,” they will gain even wider acceptance.

It is interesting to observe that both schools can be viable by ex-plaining different behavior patterns at different time frames, and assuch do not have to be necessarily competing schools. The funda-mental analysis rests on the assumption of a rational person who in-corporates all of the relevant data concerning a certain asset beforemaking a decision about its acquisition. In that regard, the past his-tory of the asset is totally irrelevant. In other words, with all the pastglory of IBM, if the fundamentals are pointing towards a dismal out-look, the investor will disregard the past history. In a sense, the ratio-nal school of thought, or the fundamental analysts, regard the valua-tion of financial assets determined by a “random walk.” In randomwalk, the prices of securities are likened to steps of a drunken sailor,where each step is independent of the previous one. Fundamentalanalysts basically assume an “efficient market” when the stock pricesreflect all information available to the public. The efficient markettheory, that the fundamentalists adhere to, assumes rationality at alltimes on the part of investors and does not allow behavior based onemotion and all other impulses. Yet as a practical matter, humanirrationality is important. Even in the legal code, the plea of insanityallows for impetuous behavior that is not based on rationality andsimply stems from a sudden urge or instant decision. Perhaps someof the most persuasive evidence against the “efficient market” theorycomes from the “anomaly” literature, which has discovered unusualpatterns in the price behavior of securities. Some of the most puz-zling price anomalies are related to seasonal patterns in the move-ment of stock prices. Other anomalies relate to returns that are de-pendent upon the size of a firm and the impact of new stock issues.

It is the contention of this author that in the short run (anything

from a day to a few months), emotions and other biases may lead usto make a decision that may not be based on rationality. Impulsivebehavior, herd mentality or any other decision-making process whichrelies on mechanisms other than rational analysis of all relevant fac-tors are not allowed in the fundamental analysis or the theory of effi-cient markets. But how else can one explain the events such as mar-kets behaving differently on Monday mornings than Friday after-noons, or that every year there is a sense of nervousness in the mar-kets around October?

One of the major problems that behavioral analysts have to face isthat in their analysis of the market they often ignore the concept ofprobability. In other words, they often sound as if they are statingtheir forecasts with certainty. As a consumer I may react to a 50%discount offer based on sudden impulse, but such impulse may notdictate my actions every time I encounter such a discount. The be-havioral analysts have to specify that their technique is only for short-run decisions and as such may be more useful to traders than institu-tional investors, such as pension plans, who are concerned with thelong term returns on assets. The technical analysts also have to finda way to incorporate probability analysis into their analysis. Other-wise, there is no basic problem with their use of past data to arrive atcertain conclusions about the future. In traditional forecasting mod-els, such as “time series analysis” such as “Box-Jenkins” the past datais also used to make inferences about the future. In fact in econo-metric forecasting, the “least square estimation” or “maximum likeli-hood” method, the forecast of a dependent variable is based on aweighted average of the past observations of the same variable. Theabove statistical methods simply determine the weights through sta-tistical manipulation, and the forecasts are based on probabilistic as-sumptions about the behavior of variables, and as such are not deter-ministic numbers. In fact, “auto regressive” estimation methods arean important part of econometrics where the past values of a variableare used to determine its future forecast.

The fundamental analysts can point to the strength of an underly-ing security based on the fundamental variables that will impact itsvalue in the future. But this analysis, by its nature, is a long-term phe-nomenon that is incapable of pinpointing the time that such move-ment will begin. In other words, the fundamental analysts can neverprovide us with the turning point.

Should we dismiss one theory in favor of another? The study ofthese two methods of security analysis reveals that they are concernedwith different time horizons and different decision-making processes.Fundamental variables can certainly affect the value of a security overthe long run. However, in the last few years, the economists haveaccepted that there are a lot of human emotions entering the pro-cess of decision making, not just calculating rational behavior, at leastin the short-run. Granting of the Nobel Prize in economics in 1993to Professor Douglas North is testimony of admission by mainstayeconomists that other modes of behavior such as culture, habit, biasand prejudice as well as impulsive or random behavior could be usedto explain consumer behavior. Increased attention paid to “behav-ioral finance” by some well-known finance scholars should open the

TECHNICAL VERSUS FUNDAMENTAL ANALYSIS:A View from Academe

Hamid B. Shomali, Ph.D.

THIS ARTICLE APPEARED IN THE MTA JOURNAL, WINTER 1994/SPRING 1995

2


door for a less-biased approach toward “technical analysis” by thetraditional finance professors. The fact that industry, such as Japan’s,has decided to invest $30 million in researching such topics indicatesthe security industry’s serious interest in the topic of technical analy-sis.

BIOGRAPHY

Hamid Shomali, Ph.D., is professor of Finance and Econom-ics, and Dean of the School of Business at Golden Gate Univer-sity. Dean Shomali joined Golden Gate University in 1986 aftera distinguished career in banking and finance. At the Bank ofAmerica, he completed several policy studies that impacted theinternational lending of the bank. Also as a member of the en-ergy-lending group, he made a substantial contribution to thebank’s energy loan portfolio. As Deputy Managing Director ofBank Farhangian, Iran, he managed the bank’s construction andmortgage lending as welt as its international operations. Priorto that he was an economist for the Central Bank of Iran wherehe completed analytical projects on a broad range of macroeco-nomic and monetary issues. Dean Shomali has served on thefaculty of several universities including the University of Califor-nia at Berkeley, University of Houston and the National Univer-sity of Iran. His teaching and research has been in internationaltrade and finance as welt as oil economics. Dean Shomali con-sults with international companies on banking, finance and in-ternational management. Dean Shomali received a Ph.D. inEconomics from the University of California in Los Angeles(UCLA) in 1973. His undergraduate education was completedat the University of Salford, England where he received a B.S.degree in Mathematics and Economics (Joint Honors) in 1968.


After many years of observations, Charles Dow concluded that thestock market, like the ocean, had three movements: primary, second-ary and daily fluctuations. The major advances and declines of themarket were equated with the tides – a dominant force that lasted fora period of time. These long-term movements were subject to sec-ondary reactions (waves) that lasted for a shorter period of time andmight temporarily appear to contradict the primary trend. And fi-nally, the waves themselves were broken down into ripple (daily) re-actions. But Dow strongly felt that no means of manipulator coulddivert the eventual course of a major primary move. It is for thisreason that he dedicated his theory to the long-term outlook for thestock market.

The discovery of the rhythmic movements of price led to the ad-vent of the Dow Jones averages. Beginning in 1897 The Wall StreetJournal published two sets of averages: the industrials and the railroads.The logic behind the specific makeup of these separate indices isrooted in Dow’s premise that both of these sectors of the market wereinterdependent. For example, if the large industrial firms of the daywere faring well, they depended upon the use of railroads to trans-port their products. Whenever the price trends showed disparatemovement between these two indices, it meant that one sector wasstronger or weaker than the other, and if allowed to continue, it wouldeventually result in a major reversal for the general stock market.

At this time in history, as our nation was still growing – pushing itsway across the entire expanse of the North American continent – theiron horse was the only means of transporting people and produce.Thus industry and railroads prospered and suffered together. As ourcountry matured and the transportation revolution took hold, theuse of railroads diminished considerably. It was for this reason thatthe Dow Theory came under recent attack. “What about the airlinesor the truckers – why are they not incorporated in this theory?” Thisargument was valid. So on December 22, 1969 Dow Jones & Com-pany revised the rail average to include other means of transporta-tion. Today, the new Dow Jones Transportation Average satisfies theoriginal requirements of a balance between industrial firms and thetransportation network.The following is an excerpt from The Dow Theory Explained byCharles B. Stansbury:

“We now come to a fundamental tenet of the Dow Theory andthat is that any signal to be authentic must be affirmed by both theindustrial and railroad averages. While this concept may seem a littleconfusing at first, we have only to return to our simile of the move-ment of the tide to clear it up. Instead of watching a single beach (orchart) we now must imagine ourselves standing at the mainland endof a narrow peninsula from which we can watch two beaches: Bothare parts of the same ocean (market) which is divided into two parts(industrial average and railroad average) by the peninsula. Whileboth beaches are subject to the same tidal action they may show vary-ing wave action. The wave action on one beach may often prove highlydeceptive as to the course of the tide unless we find the movementconfirmed by similar action on the other beach.”

The February 11, 1922, The Wall Street Journal stated:“... the stock market is acting not upon the known news of today

but upon what conditions will be as far ahead as the combined intel-ligence and knowledge of Wall Street can foresee. There are plentyof bear arguments in the complicated conditions in Europe, the un-

certainties of taxation and the interested aberrations of Congress.All these factors are known and, if possible, over discussed.”

Charles Stansbury also wrote that“Over the years during which the averages have been observed

and recorded this confirmation by both averages has established itself as anessential part of the theory.

“The confirmation which carries authority need not develop inour chart on the same day or even in the same week. It is deemedsufficient if one average follows the other into new low ground, ornew high ground, before the first average retracts its half of the signal.The first average retracts if it makes a new extreme in the oppositedirection before confirmation by the second average.”

Confirmation of Primary Bull and Primary Bear MarketsIn Chart I we depict the price trends of both the Industrial and

Transportation Averages. Note that drop B to C does not register anew low (below A). This is the first sign of a potential positive shift in

the making. At C, one or bothaverages holds above the low(A). A bear market bottom isconfirmed at point D – the firsttime that the averages pen-etrate a previous rally of sub-stance.

A bull market reversal oc-curs when there is tremendouseuphoria; the “things couldn’tbe better” syndrome. Sometime during this period, theaverage(s) is unable to register

a new high (G) above the previous and, in hindsight, ultimate peak(E). It is at point H that the primary downtrend is established – adecisive break below a previous important (real) secondary low (F).

Secondary ReactionsIn every primary move, whether ascending or descending, there

is a time when prudent investors commence profit taking (Chart II,point A). This normal process invariably causes the average(s) toweaken – perhaps both industrials and transportations show signs of

divergent trends at this point.Significant redeployment offunds becomes evident at pointD – a time when the buyers’ ac-tivity supports sagging pricesand prevents a new low fromoccurring below point B. PointE suggests more optimism asthe minor high (C) isbreached. It is not until a newhigh (F) is attained that theprimary trend can still be con-sidered in force. This entire

pullback phase (A thru F) is known as a secondary reaction (waveswithin the tides).

How does one distinguish between a secondary and a major re-versal? It is at this extremely critical juncture that one pay close at-

A DOW THEORY UPDATE

Ralph J. Acampora and Rosemarie I. Pavlick

THIS ARTICLE APPEARED IN THE FIRST ISSUE OF THE MTA JOURNAL, JANUARY 1978

3


tention to all the characteristics present during secondary reactions.To begin with, secondary declines have a multiple effect (several down-ward swings in price). In Chart III, the most recent primary upswing

began at point A. At point E theDow theoretician must come togripes with the problem of iden-tifying B through E as a second-ary reaction or a major reversal.If the verdict is “Bear Market,” hewill sell all of his or her stock andperhaps go short. However, ifthe interpretation reveals the ex-istence of a secondary reactionphase he or she will deem thisdecline a normal happening andslowly commit funds in order to

take advantage of an eventual major “up” move. Listed below arethose characteristics common during secondary phases:■ The movement is more rapid in a reversal than in a primary trend;

they may last from three weeks to three months and typically re-trace perhaps 40 percent of the movement since the end of thelast major reversal. But there have been occasions when a second-ary reversal retraced as little as 30 percent of the primary move -and as much as 70 percent.

■ The length of time needed to complete this reversal is usually amuch shorter time than the previous advance. From top to bot-tom (B thru E) more than three months time usually indicates abear market.

■ If volume during the reaction (B thru E) equals or exceeds thelevel that prevailed at the time of the top (B) it would be bearish.However, if volume continues to drop lower as prices decline thereis a good chance that this decline is nothing more than a reac-tion.

■ The atmosphere surrounding this entire period is also very im-portant. If excessive speculation is present then this type of reac-tion could be interpreted as the beginnings of a bear market.

“...The market barometer does not pretend to do the impossible.It forecasts, defines and confirms the major swings... It does not pre-tend to forecast the secondary reactions any more than it clearly fore-tells the corresponding rallies in a major bear market.

“This is because the secondary reaction, as distinguished fromthe major movement, is governed by the unexpected.”

The Wall Street Journal, September 19, 1922It is now imperative to define the distinctive phases present in

Dow’s bull and bear markets:

The Bull Market■ Phase One is known as the accumulation phase – very depressed

prices – basic industry, utilities and high yielding stocks dominate.■ Phase Two is characterized by increased activity, rising prices and

an improving business scene. Secondary stocks are in vogue.■ Phase Three, the final explosive move, a by-product of excessive

public speculation. The ‘cat and dog’ syndrome.

The Bear Market■ Phase One is called the distribution phase. As the “public”

scrambles for stocks, the farsighted begin their deliberate selling.Stocks go from strong to weaker hands.

■ Phase Two is referred to as the panic phase. Selling begets sellingas the urgency to liquidate mounts. Margin calls escalate.

■ Phase Three is marked by continued erosion in prices, the crunch-ing of lesser quality issues – object pessimism prevails. “It's alwaysdarker before the dawn.”

THE THEORY TODAY

Charles Dow specifically emphasizes the use of the closing pricesfor both averages, because he felt that these figures would give a truepicture of the floor traders’ and specialists’ positions. Despite theirlong and short dealings, during the day, these professionals wouldinvariably even up before the close.

On Chart IV are depicted graphs of both averages: we have in-serted the closing figures of the key reversal points and non-confir-mation levels since September 21, 1976.

Dow Jones Industrial Average

Dow Jones Transportation Average

Chart IV


This entire period is in question. Are we in a secondary reactionphase or has a Dow Theory bear market signal been given?

In determining a trend, previous high and low points are used. Asuccession of new highs is positive while a series of new lows is nega-tive. The Dow Jones Industrial Average on Chart IV has been tracingout a progression of lower highs since September 1976 (note pointsA, D, E, G, H, J and K). Points B, C, F, I and L vividly portray asuccession of lower lows – this combination is negative.

When viewing the Dow Transportation Average from October 1976to May 1977, a distinct divergence is seen. Points. P, S and T repre-sent a classic series of new highs while N, Q and R are importanthigher reaction points - this combination is positive. The DJIA reachedits high on September 21, 1976 at 1014.79 (A) while the DJTA regis-tered its high on May 18, 1977 at 246.64 (T).

Remember Dow’s fundamental tenet: “for any signal to be authen-tic, both averages must confirm. It is deemed sufficient it one aver-age follows the other into new low ground, or new high ground, be-fore the first average retracts: its half of the signal.”

Now the question is raised – since the Transportation Average hasrecently come under sharp selling pressure (points W and X), doesthis move constitute a confirmation of the Industrial Average’s nega-tive behavior?

To begin with, the Dow Jones Transportation Average was con-fined to a tight trading range during the months of June and July1977 (U and V). This is called a line formation. It usually lasts sev-eral weeks with price fluctuations in the magnitude of 5%. Suchmovements indicate accumulation or distribution. Any break belowthis formation is distribution and implies lower prices. Needless tosay the Transportation Average has decisively penetrated the lowerend of this pattern (W); in so doing, it also weakened below its Feb-ruary 25 close of 221.81 (Q). It is here that the Dow theoreticiansdiffer. Is 221.81 the critical secondary low? If so, then the DJTA hasconfirmed the breakdowns in the Industrial Average and has initi-ated a primary bear signal. A secondary low must be considered thebeginning of the recent primary upswing. We interpret the Oct. 3low (N) of 203.85 the focal point for the primary upswing. Thus,until the Transportation Average closes below 203.85, the move downfrom T is still considered a secondary reaction.The following excerpt from Richard Russell’s The Dow Theory To-

day is noteworthy:“Bear market signals, however, must not be oversimplified. The

great Dow Theorist Robert Rhea wrote in 1938: ‘Beginners frequentlymake the mistake of basing conclusions wholly on the matter of pen-etration. Familiarity with the correlated factors such as duration,extent, activity, divergence, and secondary implications of primarybull markets is needed to make a correct diagnosis.’ Anyone who hasstudied the works of Hamilton and Rhea knows that it is only in thethird and last phase of an extended bull market that a bear signal isvalid. Ignorance of this fact has led to one of the most disastrousmisreading of the Averages in modern stock market history.

“Over and over again the great Dow theorists have warned us notto take a shallow, mechanical reading of the Averages while disre-garding phases, duration and extent of the market movements. Bycalling a bear market on a ‘false’ second phase signal, the majority ofthe financial fraternity has committed one of the most costly errorsin market annals.

“Once the fact is accepted that bear market signals are valid onlywhen they occur within the third phase of a bull market, the utmostimportance must be attached to identifying the third phase. ‘This isthe time,’ wrote Rhea, ‘when brokers and soothsayers prosper, andwhen an excited public, lured by the bait of advancing prices, buysstocks without regard to values; basing their action on nothing morethan hopes and expectations.’ He observes that ‘this is the phase

where worthless stocks are bought for no other reason than becausethey look cheap and because gamblers hope they will double in price.This condition always has prevailed in the third phase of bull mar-kets. . . .’”

Let us now investigate Rhea’s co-related factors:■ Duration: The Dow Industrials has easily exceeded the three-week

to three-month time limitation used in measuring a secondaryreaction. The Transportation’s reaction began on May 18 (T),also overstaying this requirement.

■ Extent of the Decline: A secondary reaction typically retraces per-haps 40% of the movement since the end of the last major rever-sal. On occasion, retracement of 30% or as much as 70% of theprimary move have been noted. The DJIA has retraced 34% of itsprimary advance that began in December 1974. The DJTA hasonly retraced 26% from its primary low, registered in October1974.

■ Phases: To date we have witnessed impressive moves in the basicindustries, utilities, and high yielding stocks – the fulfillment ofPhase One in a bull market. Secondary stocks have moved to thefore and dominated the scene in the past twelve months, thus sat-isfying Phase Two. However, the overheated, speculative fever stagehas not been witnessed; thus the paramount requirement has notbeen met. The reactions to date (A to L and T to X) have nottaken place during the third and final phase of a bull market.In conclusion, “it is only in the third and last phase of an extended

bull market that a bear signal is valid.”

203.85 This number represents the important secondary low reg-istered in the Dow Jones Transportation Average on October 12, 1976.Since September 1976, divergence has existed between the DJIA andthe DJTA causing many Dow theoreticians to question the viability ofthe bull market that began in January of 1975. Last Monday, theDow Jones Transportation Average closed below 203.85, giving a bearmarket signal. However, everyone was fully apprised of this develop-ment – leading Wall Street publications contained articles describingthis phenomenon and its ominous consequences. The nonbelieversquickly responded with either a shrug of the shoulders or the state-ment that this signal was “too much and too late.” Some sellingcame to the fore because of the breakdown but quickly reversed intothe first 5% rally in 1977. That rally nevertheless does not negate theimportance of the signal. In fact, the rally comes as no surprise – themarket had already suffered slow deterioration for several months,and at the time was in an extremely oversold condition. The reac-tion was anticlimactic, but don’t be fooled by the resultant rally. Muchmore is needed to reverse this negative signal. Charles Dow com-pared the market’s primary trend with the ocean’s major tides. Hesuggested that within the tides, waves would occur (secondary reac-tions) that move counter to the major flow. These counter movescould extend from three weeks to three months and have no lastingeffect on the major trend. Thus, if this bear market signal is accepted,temporary rallies could be viewed as a selling opportunity within amajor downward phase.

Written 8/19/1977 3:30 pm DJIA 862.27


There has been a steadily growing interest in Contrary Opiniontheory over the past few years. Beginning in the 1940s, Humphrey B.Neill, who died in the mid 1970s at his homestead in Saxton’s River,Vermont, wrote about Contrary Opinion in his now retired Neill Let-ter of Contrary Opinion. I joined him in furthering the essence ofthe theory in 1962 with The Contrary Investor, a newsletter on in-vestment implications of Contrary Opinion that I continue to writetoday. Moreover, I began to reprint old books that deal with humanbehavior and the stock market while slowly moving into money man-agement utilizing a Contrarian strategy.

Today, Contrary Opinion is accepted as an investment tool and,in fact, has become part of conventional wisdom. Whereas for yearsthe uses of Contrary Opinion were always in the back room, now themainstream has recognized that to make money you buy the down-trodden, the misunderstood and the overlooked. Also, there arenumerous investment letters, books, and practicing managers whoseek lesser recognized or secondary growth stocks which do havegrowth characteristics but not growth multiples. New investmentbooks have contrary or contrarian in their titles and institutions, nowresponsible for perhaps 85% of stock market trading, nod sagely atcommittee meetings when a manager says he uses Contrary Opinion.

However, reality is not that pure. Many people use the words butnot the strategy. Long-term Contrarian investors, which means valueplayers, include Warren Buffet, John Templeton, John Neff, DeanLeBaron, Phil Carret, Ir ving Kahn, and David Dreman. They allmanage significant sums of money and have done so for a number ofyears. Of course there are others, but at least this gives you an idea ofwhat I mean.

Phil Carret said years ago in his book, The Art of Speculation,published in 1930 (he is still alive today managing money in NewYork at the age of 89) that the road to success in speculation is thestudy of values. “The successful speculator must purchase or holdsecurities which are selling for less than their real value, avoid or sellsecurities which are selling for more than their real value. The suc-cessful investor must pursue exactly the same policy.” Of course, thetime requisite for prices to move up is more important to a specula-tor than an investor. “A security may be undervalued, but if it is alsoout of style it is of little interest to the speculator.” So, one has tostudy the psychology of the stock market as well as the elements ofreal value. When real value is out of favor, a Contrarian moves in. AContrarian investor waits patiently. A Contrarian speculator, on the

IT PAYS TO BE CONTRARY

James L. Fraser

other hand, tries to judge the psychological climate with other tools,as charts and technical indicators that will allow him not to wait toolong. Let me give you an example. The chart below from M. C.Horsey & Company shows Sears, Roebuck, a major American corpo-ration. Just before the August 1982 rise began, Sears was ourcompany’s largest holding. How did we get that way? Sears was anifty-fifty stock back in 1972 and early 1973 and then declined withthe bear market of 1973-74. It recovered in 1975 and early 1976 andthen sank into its own tedious bear market.

During this time, the news was largely negative in that merchan-dising was not doing well, other firms were taking market share awayfrom Sears and as each year went by the financial press reported moreand more negative news. The price kept coming down until finallynegative news no longer pushed it down. The stock began the bot-toming process in 1980 and, as long-term investors, we began buyingin late 1980 right up to late May in 1982. We felt we had value whichwas then not being recognized.

As long-term investors, we bought too soon but that was not verymaterial once the stock moved up strongly in 1982 and 1983. How-ever, a speculator would have timed the movement better and per-haps bought after a breakout above 21 in 1982. The main Contrarianpoint is that once negative news no longer pushed the price down tonew lows, the price fairly represented all possible disappointments.Of course the chart looked terrible at that time as past history for tenyears was downhill, but the unexpected income of positive change,which takes a long time in a corporation as large as Sears, was aboutto be the next major factor which coincided with a market move thatbegan in August 1982.

Take another example of a recent out of favor company with thechart of Halliburton below. Everybody is aware of the great movethat took place in energy related securities during the 1970s that, inmost cases, peaked out in late 1980. We then had a sharp fall back, arise that participated with a 1982-83 bull market, then a continuingdown turn based on negative fundamentals for major industries withinthe energy area. Oil field services represent a volatile sector.

Finally a Contrarian is attracted to the stock price after seeing itdecline significantly from a high level, keeping in mind that the highprice represents an extremely optimistic scenario and one that wouldnot last. The question becomes where is value and in relation towhat level of oil prices. An investor might begin buying in the low20s and certainly participates below 20. A speculator or trader would

4THIS ARTICLE APPEARED IN THE MTA JOURNAL, NOVEMBER 1986


wait for more price confirmation, that is of the price stabilizing wherebad news is no longer a factor and where perhaps the next level ofnews is likely to be favorable.

PREDICTIONS AND FORECASTING

Fred C. Kelly, a writer for the Saturday Evening Post back in the1930s first published Why You Win or Lose in 1930 with a for ward,interestingly enough, by John B. Watson, the founder of behavioralpsychology at Harvard. This book is a favorite reprint of ours and inmy preface I say we have difficulty struggling against crowd behaviorpatterns. We don’t compel thinking and observation. We don’t workat contesting the popular view. We are mislead by financial propa-ganda that makes us untimely in our opinions and often wrong inour actions.

Kelly understands that human gullibility's are a constant contribu-tory force to speculation. The only change between Kelly’s days andour own time is that crowd reactions occur faster, thus opinions shiftquicker, and jumping to conclusions becomes an unprofitable pas-time.

Every natural human impulse seems to be a foe to success in themarket. We all want to conform, to congregate as a herd. And yet wewin by understanding human psychology and by thinking ahead increating possible courses of action to today’s conformity. You don’thave to be a highly competitive mental person to succeed, you onlyhave to watch and study the crowd in order to pick up useful clues asto what the intelligent minority is not doing.

To succeed in the market one must not do what most others aredoing. He who does the opposite has a good chance to be right. Wemay not know what insiders are doing until after they have done it,but by watching and studying the investing crowd we pick up clues asto what they are not doing. I do not mean that you want to avoid amajor current when it is strongly flowing because the uses of Con-trary Opinion are most valuable at turning points. To get aboard amajor move at the right time, it is only necessary to disagree with theopinion of most investors you know who follow logical reasoning pro-cesses fostered upon us by print and TV financial media.

It is not the stock market which beats us. It is our own unreason-ing instincts and inborn tendencies which we do not master and which,when we give in to them, lead to disaster. Natural instincts governaction which means that fear and greed are at the opposite end ofthe investment spectrum. I know, we all think, times have changed,which they have, but human nature has not changed and that is thepoint of wisdom regarding Contrarian investing strategies.

Another way to look at this is to say that money may be managedprofitably or conventionally but not both. Of course, stock prices rep-resent consensus expectations. But to profit, the expectations haveto be both correct and different from what is current conventionalwisdom. A significant human problem is to withstand group forcesthat seek to modify and distort individual judgments.

The business of investing is an actual study of social influence.Personal investing is so widespread that a social group has beenformed, and we, as individual investors, are no longer indifferent tothis group. When we visit brokers’ offices we are alert to the group.If we hang around long enough, we tend to reach an agreement withthe prevailing opinion since this is the dynamic requirement of agroup situation. Otherwise, our personalities would suffer. Evenwatching business news on television puts us in the position to acceptprevailing opinion. The antisocial solution is to turn off the set.

TREND IS NOT DESTINY – CHARACTER IS DESTINY

We tend to modify our judgment in response to the pressure ofmajority and expert opinion. Most financial media reinforces major-ity beliefs and convictions. Investment practices are adopted on thebasis of reasons that appear valid. But each investor comes underthe sway of an already existing system of practices and values so hecannot judge independently, and he is affected the most when he isleast able to exercise his own judgment. In other words, characterand temperament are more important than charts and systems. Wewant to endow investing with specific guides that can be counted on.This is good. But we should realize that our emotions and uncon-scious behavior patterns, as the tides of the Bay of Fundy, often over-run these guides and just as often leave them stranded.

Independence and basic confidence in your ability to controldoubts is a primary requisite for successful investing. Take the caseof Xerox as represented in the chart below. Obviously, when Xeroxwas above 150 in 1972 and 1973, there was little independent think-ing regarding the idea that the stock deserved such a high price.This was group behavior in all its splendor. Then came the marketdrop of 1973-74 with a bit of a recovery thereafter, but not muchwhen you consider the market bottom in recent times was year 1974at 570 on the Dow, and the market has moved higher since then.What happened is that the majority belief and conviction that Xeroxwas a special company, deserving of high multiples and all good things,was dashed through the rest of the 1970s and into the 1980s. To besure, the higher you were in 1973, the easier it was to fall down andhurt yourself.

Anyway, right through 1982 and into 1984 the majority belief be-come one of Japanese competition winning the day with copiers, firstat the cheap end of the line and then finally across the board. Xeroxmanagement was perceived to be incompetent and not paying atten-tion to what was going on. We bought the stock in 1982 and again in1984 an the basis that management, though a bit thick, was slowlyresponding to the global changes influencing its business. Thesechanges take time. Finally, by 1984, negative articles no longer influ-enced the price of the stock. That was the safe time to buy in whattechnicians might call a saucer bottom. The stock then became ourlargest holding.

Interestingly, as you might suspect, as the stock rose up, we finallybegan to see a few nice words about Xerox. News follows the price. As the price moves higher the financial media speaks sweetly.

PRIDE OF OPINION PRECEDES A FALL

I have been saying that a practicing Contrarian observes the psy-chological status of the crowd in question and then takes an oppositeapproach. Of course, there are more than one opposite approaches.Normally, if the crowd has decided upon a conventional future, thenthe successful opposite approach is either more positive or more nega-


tive, at the extremes, until one of the extremes becomes the conven-tional view and then the Contrarian again must take a different roadfrom the conventional view. The theory is based upon estimating theprevailing crowd emotion and not on forecasting the future. We makeno attempt to predict the future, and we keep an open mind. Anydefinite forecast of the future leaves one at the mercy of that forecastbecause pride of opinion will tend to tie us down to that forecast.

Take a look at a medium-sized stock chart, which is Ransburg be-low. This company is not followed by Wall Street very closely, is lo-cated in Indianapolis, Indiana which people may come from but rarelygo to, and reflects future forecasting in price moves. Not of the com-pany itself but of fields of activity that the company represents. Basi-cally, Ransburg is a leader in industrial equipment and is associatedwith robots. Not surprisingly, the all time high price of above 37 wasmade in April 1981 when Time magazine had on its cover robots andthe tremendous future they represented. This cover journalism fore-cast of the future left investors at the mercy of brokers who were push-ing robotic stocks. Anybody who read Time wanted such items andRansburg, normally a quiet stock, topped out at the peak of enthusi-asm for robots. Obviously, this enthusiasm was not limited to themonth of April but had been building up since the second half of1980.

Again, the high technology boom in 1983, which maximized inJune of that year, also saw Ransburg peak out on the basis of high

technology and robots.The price came downagain only to rise a bit in1986 on the basis of in-dustrial expansion andthe improving of Ameri-can productivity. Nowthe stock is back downonce more to below the15 area and is a ContraryOpinion buy since we canchoose it with an openmind that is not influ-enced by delightful fu-ture forecasts. Theseforecasts will come to lifebut not when anticipatedand so far we have had

three false moves which undoubtedly will keep investors away untilthey miss the real move that takes place one of these days.

Oh, I realize that it is easy to make points from charts that backme up. And yet, investing is not a science, and one must enjoy it todo it well. However, we should stick to what we understand, havesome guts, and never become too overconfident. Again, Phil Carretonce said something to the effect that an investor who so lacks confi-dence in his own judgment that he won’t buy any security until it isfavored by the consensus of the investment community, will buy fewbargains and is unlikely to achieve superior results. Of course, a se-curity favored among professional investors is good but that is notthe same as being undervalued in price. If unpopular, or generallyunrecognized, some investigation is required to estimate basic value.That is where security analysis comes in. Once assured the basic valueis there, then unpopularity will not deter the investor who is 1ookingfor long-term results.

A good trader, speculator or market technician is trying to do thesame thing though using somewhat different tools that offer insightsinto value. We are all trying to buy things where the future is notalready discounted. We want stocks with merit and we should buy

them when they are weak. But usually we buy stocks with merit whenthey are strong and thereby do not build good performance overtime.

PATIENCE IS SUSTAINED COURAGE

Another way to look at this is at perspective and patience of theessential requirements. We all want to own shares of successful com-panies in areas of activity that have particularly promising future pros-pects. However, the Contrary lesson is that we all tend to be influ-enced by whatever feelings are sweeping over the investment com-munity at the moment and that true investing, to be successful, re-quires fighting these feelings. Nevertheless, it is not that simple forthe inexperienced investor to be contrary since inexperience breedsa certain contempt for long-term solutions.

Subscribers to advisory services aim at quick results, feeling thatthe game is not worth the candle unless a system or technique worksimmediately. An individual, fortunate enough to have an intuitivesense of values, should be able to achieve reasonable profits with somedegree of consistency. The key words here are reasonable and con-sistency – words not in the vocabulary of those who do not yet havemarket experience.

R.W. McNeel, Financial Editor of the Boston Herald from 1912-1922wrote Beating the Stock Market, published in 1921 on the humanside of speculation – which means attitudes, beliefs, hopes, and fears– the emotions and characteristics that any of us associate with hu-man beings. Now studying people may not seem rewarding. But ifyou subscribe to the thesis that confidence makes business you studypeople. Writers tend to emphasize their statistical figures, but peoplegive these figures meaning.

My point is that character is as essential as knowledge, even moreso today when basic statistical knowledge is readily available. InMcNeel’s day balance sheet figures were less reliable and yet theywere relied upon. In our day statistical analysis is clearly stated andasset values are known yet the same stock swings take place. Theconstant element is the human side of finance – that has not changed.Natural instincts will unquestionably govern action.

To illustrate, fear is probably the oldest human instinct. It is uni-versal and deep rooted. It is the outgrowth of self-preservationwhereby we have been able to survive over time. To quote McNeel:“Because of its ancient origin and its great strength, man is at timesexposed to the absolute breaking down of his courage under certainconditions and frequently without cause.”

The other side of fear is greed or from the instinct point of viewthat of companionship or gregariousness. Investors tend to flocktogether. We have an inborn tendency to do what we see someoneelse doing. Investors become excited and tend to act like lemmingsas they enter into the active or emotional states of others.

Consider how basic instincts cause us to act in the stock market.We are told to buy low and sell high. Yet stocks are never low unlessthe headlines are such as to cause the great majority of active inves-tors to sell. Selling is usually a creation of financial necessity, need-ing money, or more largely through fear of pending developments inthe world that make you feel that prospect of financial 1055 is cer-tain. Stock market bottoms are created when we sell stocks at ridicu-lously low prices without conscious reason. Whatever that price levelis at, it will look low in retrospect. Recent examples are Dow 570 in1974, 770 in 1982, and 1079 in 1984. This may be a pattern of risingbottoms, but each one offered exceptional opportunities for the pur-chase of stocks. The opportunity is only there because most of us areunable to turn our instincts or emotions upside down and buy whileothers sell.


McNeel said: “In order to avoid selling, and on the other hand tobuy, he must put his natural inclinations to the test of reason anddetermine whether they are sound or unsound.” Every investor triesto do this, and I find from my own personal experience that it iseasier to do this if you are physically divorced from the financial blackholes of enthusiasm or despair which means the major metropolitancenters. I am in Burlington, Vermont because it is just enough offthe beaten path to make it a Contrarian’s delight. Hopefully, thiskeeps us a bit away from the inborn tendency to act in common withothers.A few axioms to the uncertain art of economic prophesy follow; I will

create a baker’s dozen guidelines:1. The system is there is none. I know this sounds strange but suc-

cessful techniques are counterproductive when widely followed.Systems that work usually come to your dinner table as food forthe future when actually they are the result of previous activitythat is now so popular the system is not worth buying. What oneneeds are systems at the breakfast table, for sustenance over thecoming day. Usually, one receives a system too late.

2. Consensus hopes or fears are embodied in current valuation lev-els. The corollaries to this are that realization of expectationsresults in no price change while realization of unexpected out-comes moves prices. The chart of Aetna illustrates this point ofview which is represented by “The Trader” column by Floyd Norrisin the 18 June 1984 issue of Barron’s [see Exhibit No. 1]. Norrishad attended the seminar on Value Investing sponsored by theNew York Society of Security Analysts where I was a speaker. Hisattention was drawn to my mention that property/casualty insur-ance companies appear cheap in the marketplace.The letter to the Editor is from the 25 June issue of Barron’s andconcerns our mentioning of bottom fishing for property/casualtyinsurance companies [Exhibit No. 1]. The point of Mr. Swift’s com-ments is that the entire industry is so bad off that our timing isnowhere near correct and investors should continue to avoid theentire area.Of course, Mr. Swift’s letter represents good thinking and its in-clusion in Barron’s reflects a certain style of consensus wisdom.The question comes down to that once your worst investment fearsare realized, there is only upside potential left. But what are theworst fears? Is there a real crisis in the property/casualty insur-ance group or is it a case of the negative atmosphere being sostrong that this is not the time for selling but rather the time forbuying? Needless to say, the stock prices of both Aetna and Con-tinental were both within 30 days of lows that have not been seensince.

3. The ability to sense what is going on in the economy is more im-portant than organizing facts. The result is that indicators are nosubstitute for judgment. Besides, a simple yardstick of value canbeat exhaustive consideration of all relevant facts. When you havetoo many facts there is the question of selection. To illustrate, lowprice-earnings ratio investing is an extremely simple strategy whichworks over time, probably because of its simplicity.

4. There is a failure to perceive new reality. In other words, it isdifficult to see the significance of outside events which producenew watersheds. A recent example is the climb of energy pricesinto 1980 and their subsequent decline into 1986. Most of us actlike generals who are fighting the last battle in a new arena wherewe carry our experiences forward without taking into consider-ation the changing environment. The future is not always a con-tinuation of the past. Be skeptical of past trends being stretched

far beyond the present. The elastic may break or snap back whenleast expected.

5. There is a cultural-psychological lag in experience over expecta-tions. We all have a tremendous capacity to believe anything untilit is no longer worth believing. The human factor tells us that wesee things according to our preconceptions, which then paralyzeperceptions. Indeed, when the market operates correctly, eachinvestor reports a slightly different version of what is going on inthe market and what signals he feels the market is sending us.

6. Trend is not destiny. The future is never clear, and one pays ahigh price for a cheery consensus. The result is that uncertaintyis a friend of the long-term buyer, and we need to take a positionwhen most lonely. Easy to say, but not easy to do. Be suspicious ofwidely held views. Educate yourself for ambiguity instead of cer-tainty, and you have a chance for success.

7. Character is destiny. Do not put your trust in those who are try-ing to hustle you but rather believe in your own common sense.Remember that our behavior patterns are restless and dynamic,with emotions often making for strange statistical measurements.For a Contrarian it is better to be right by oneself than be wrongin good company.

8. The bedrock of reality is a world of disequilibrium. Instability is afundamental characteristic of transition. As Le Bon, the Frenchwriter of The Crowd says, a crowd yields to instincts that individu-als suppress. Rational outside perspective tends to remove youfrom current investment climates.

9. The art of forecasting is in the choosing. One has to decide whatis important and what is not. We need time to reach conclusions.We react to each new piece of information as concrete evidenceof a new trend that supports some exciting premise. The rise ofgold and silver in 1980 was a wonderful popular delusion. TheHunt family of Texas was then worth over $5 billion and now, af-ter some years of adversity, the family fortune seems to be wellunder $1 billion. Still, a nice piece of change but not exactly whatthey were used to in 1980.

10. The future is both promising and threatening. This is always thecase whether or not perceived to be. So, how you look at thefuture is largely dependent upon your own psychological makeup.Every human is anxious. We want to hear answers we agree with.

11. It is better to know some of the questions than all of the answers.That is because the answers don’t work out. To be sure, things aremore like they are today than they ever were, but most of us worryover answers to an extent that worry consumes the spirit of ac-tion.

12. Complacency breeds surprises while fear breeds opportunity. Wehave a few dozen buttons with sayings on them, some of which arepart of this article in various sentences. In this regard, patience issustained courage, and without curiosity, conviction is stubborn-ness.

13. Take not thyself too seriously. Take your work seriously but don’tconfuse brains with a bull market.This is a good place to say that it is of major importance in using

the Theory of Contrary Opinion to be contrary to words and opin-ions, not to facts. It is words that mislead, distort, and delude. Toparaphrase Gustave LeBon, one of the great writers on the crowdmind, we see how words are used as a mechanism of persuasion. Thefour requisites are:■ Affirmation – affirm the word as truth■ Repetition – repeat over and over


■ Contagion – finally it catches■ Prestige – and imitation results

Contrary thinking cannot advise you – it can only suggest Con-trary trends. We are creatures of habit and prone to recession orprosperity mentality. The irony is that widely followed forecasts bringabout their own demise.

Furthermore, one does not get by in being contrary alone, onemust also be curious. Nobody can afford to jump to conclusions sincewe have to look at the whole picture. The Contrarian can clarifythought so that recommendations fall into place. But don’t forcethem. Think first. Don’t expect the market to conform to your ownpreconceived opinions. It won't. Flexibility plus thought plus workequals a chance for success. Rigidity of mind does rot.

OBJECT, PURPOSE, METHOD, PREMISE

Let us review a bit and put more pieces together. The object ofcontrary thinking is to challenge generally accepted viewpoints onthe prevailing trends in politics, socio-economics, business, and thestock market. Opinions react sharply as people’s emotions – theirhopes, fears, and passions – sway back and forth.

The contrarian’s purpose is to contest the Popular View becausethe view is usually untimely, misled by propaganda, or plain wrong.

The method is to compel thinking and observation in place ofconclusion hopping and snap-guessing. “Think prodding” or the nec-essary concentrated reflection takes practice.

The premise is that alternative ideas make for a clearer, better-defined judgment. Taking a contrary position frequently will sug-gest what is NOT coming next. When thinking through opposites,one is led to sound thoughts as to what might come next.

Is the public always wrong? Is the Crowd never correct? After all,we live in the most enlightened democracy of contemporary times,and individuals – acting in mass – pull the voting levers.

For a correct answer we must rephrase the question. Is the publicwrong all the time? No. The public is probably right more of thetime than not. But the public is right only during the trends andwrong at both ends – usually wrong when it pays to be contrary. (Ifeel we may include institutions as acting public-like in their invest-ment activities. The market sheep are not all individuals. The fatterflocks just trample more ground.)

Professor H. F. Harding of Ohio State once wrote me that “theodds are always good that the exceptional man is well ahead of thecrowd. When they catch up he is off in another direction.” Themodern problem is that the speed of change influences the processand tends to compress all movements. Remember Voltaire who said:“It is only charlatans who are certain. Doubt is not a very agreeablestate, but certainty is a ridiculous one.”

Contrary Opinion theory is being discussed more and more inthe financial media as both professional managers and amateur indi-viduals tend to use the expression when it fits them. Accordingly,everybody is becoming aware that to do the opposite of what mostpeople are doing is the way to win. Simply, you win by being con-trary.

Most of us should act as true investors. Forgetting about marketliquidity and the speedier ticker tape, while relying on our judgmentsof what an investment will bring us over longer periods of time thana trading cycle.

There are different approaches to the problem. The specific in-vestment need of many Contrary Opinion readers is to tailor a pro-gram around undervaluation. This means you establish that securi-ties purchased are worth more than they are selling. Characteristicsand criteria are set up. The following fundamental guidelines serve

as a beginning:1. Past records give a point of departure for analysis. Average earn-

ings, dividends, asset values and their trends should be examined.Tangible value is the secret, either in a turnaround situation or aspecial asset stock.

2. New and relevant facts that expect to have an influence may bepresent. These facts should not yet be fully realized and appreci-ated by a majority of the financial community. (Clearly, technicalanalysis offers portraits of sentiment which aid the decision mak-ing process.)

3. A lower speculative component is essential. The measurementand delimitation of securities into investment and speculative ar-eas is desirable. The method is largely to ignore popular trendsand to buy ex-public participation.

4. After basic principles, the distinction is still one of personal imagi-nation and ingenuity. Confidence in market level factors influ-ence price-earnings multiples. But a strict ladder analysis, whereyou try to escalate your stock over comparative choices, is not goodpractice.

5. Try to purchase under favorable conditions. A clear-cut demon-stration of superior attractiveness is still subjective judgment. Factsand ideas favorable to purchase are remembered, while negativefactors are forgotten.Do not adhere to any formula or system. Keep no idols, but rather

stoke your noggin with antidotes for the temptation of conformity.Rely not on a consensus indicator approach as a substitute for com-mon sense. Then you will not be short-circuited. In fact, you mayeven win.

BIOGRAPHY

James L. Fraser, president and founder of Fraser Management(309 S. Williard St., Burlington, Vermont 05401), is responsiblefor overall investment guidance as well as fundamental economicresearch and portfolio management. He is a Chartered Finan-cial Analyst and has been an investment counselor and financialpublisher since 1962. He is also a member of the Market Tech-nicians Association.


Swift, 25 June 1984 in Barron'sNorris, 18 June 1984 in Barron's

Exhibit 1


5Any investor who thinks, “a bull market is a bull market, is a bull

market, is a bull market” does not understand bull markets. Thereexists a very special type of bull market, a “Runaway Bull Market,” orRBM. Because I know of no literature on RBMs, I would like to pulltogether some of the conclusions about their characteristics.

The most difficult challenge of all, of course, is the very act ofidentifying this special type of bull market. (That brings to mind thevignette about how to grow a great English lawn: the gardener de-scribes using the greatest seed stock, the finest loam and fertilizer inthe soil, the need to tend it carefully every day, to have good loca-tion, and so on. And finally the punch line: one needs to do all thesethings for 300 years.) Comparably, I spotted the RBM based on manyyears of experience. In fact it is more in the fulfillment of a set ofcharacteristics that we know a RBM is in force, rather than in somespecific signal. The young man knows the rules, but the old manknows the exceptions.

Here are the identifying features that I feel delineate a RunawayBull Market:1. A wholesale disregard of classic Technical Indicators; not just a

few, or many, but where nearly all negative Indicators are wrong.Indeed, the RBM of the late eighties has shrugged off such hoarynegative Technical Indicators as Stovall’s Four-Month General Mo-tors Rule, Gould’s Three-Step-and-Stumble Rule, the Inverted YieldCurve, “overbought” markets, and more.

2. Setbacks are either aborted or are too small and brief to trade. Inother words, after a sharp rise, instead of the typical “1/3 to 2/3Technical Correction” such as that first identified by Charles H.Dow at the beginning of this century, a RBM will have virtually nosetback. Instead, they have rolling adjustments within flat tradingranges that simply amount to preparation for the next surge ofbuying.

3. One would logically expect bad economic news to send the mar-ket lower, but RBMs seem to be stimulated by bad news to betterperformance! For example, automobile sales dropped during1989, yet General Motors continued to make new highs. In fact,negative economic numbers can actually trigger buying frenzies.A RBM either has no perceivable reaction to bad news (1) say, anoncoming recession (2) or, perhaps, it dips slightly but is thenfollowed by explosive buying surges.

4. There is no stock market more profitable – or dangerous to over-stay – than a RBM. Since RBMs tend to run their course like the“young blood” of youth, this is a great time for call options: stocksjust seem to go up, up, up. Research suggests that RBMs frequentlyoccur after great economic prosperity, especially during the specu-lative finale in low-priced stocks that usually occurs near the endof long bull markets. During a long prosperity, profits have timeto filter down to the lowest common-denominator companies,where a small profit has a huge impact on the bottom line. Thus,during a RBM, especially in its later phases, one would expectmore bullish action on the ASE and OTC markets than on theNew York Stock Exchange. If, it is mostly blue-chips on the NewYork Stock Exchange that are in the limelight, we know that thereis more time ahead of us before we need to sell.

TEN WAYS TO PROFIT FROM A “RUNAWAY BULL MARKET”James Dines

5. When a RBM ends, it tends to provide a spectacular opportunityto make money on declining markets. In other words, a RBM isapt to end in a “spike” rather than a leveling off to form a gradualTop, so selling must be executed with great precision. RBM’s areincredibly profitable and well worth playing, but very close atten-tion must be given to stop points. You should instruct your bro-ker to act accordingly as soon as a trigger point is reached.How to detect the Top, or sell point, therefore assumes paramountimportance, especially considering that we cannot rely on the usualTechnical Indicators.Are we merely reduced to selling whenever an Uptrend line isbroken, or when a stop-loss point gets triggered? No, but the maintechnique will be in the psychological realm. Therefore you mightwant to review the first third of my first book Technical Analysis,which covers mass psychology, especially the “Greed/Fear Oscil-lation.”

6. By far the most prominent characteristic of a RBM is that the in-vesting public and professionals regard it with profound skepti-cism and disbelief. This is a crucial element, and there can be noRBM without it. Thus, we are led to consideration of how we canjudge when public fear is replaced by greed, for that will be thefirst sign of the RBM’s end. Focusing on that crucial aspect ofRBMs, we first must distinguish between stock market and eco-nomic factors.The Conference Board reported that consumer confidencereached a 20-year high in July 1989. Psychologically, that was be-cause consumers are looking at low unemployment levels around5%. Also, with around 200,000 new jobs being created each monthin the US, dropping interest rates, and low inflation levels, suchconfidence was understandable. But note very carefully that con-sumer confidence has little to do with stock-market investor con-fidence; the latter’s motivation is fear and a clear, sharp memoryof the Oct 1987 bear market (or “crash,” as tyros might describeit).For what then should we look when measuring the type of publicpessimism that would keep us in this stock market-hopefully mak-ing big profits? Most important is what is NOT happening! Forexample, when the Dow Jones Industrial Average (DJI) made anew all-time high in August 1989, there were no magazine coversfeaturing this event; in fact the new high was greeted by yawningindifference! Also, there was no news of “hot new issues going toimmediate premiums.” There were no large secondary offerings,and in fact the news emanating from Wall Street was remarkablysubdued considering the DJI’s huge rise since 1987. Because thepublic buys its stocks through Wall Street, Wall Street is a wonder-ful mirror of what is truly happening in the minds and wallets ofinvestors. Note carefully it doesn’t matter whether investors talkoptimistically, only whether they actually buy stocks. During 1989the cash position in mutual funds was increasing steadily. In otherwords, fund managers sold into the strength because they disbe-lieved it!Yet more evidence of professional pessimism at that time couldfigure in the price of a seat on the New York Stock Exchange. In

THIS ARTICLE APPEARED IN THE MTA JOURNAL, WINTER 1989/1990


late 1989, the price was below where it had been in Oct 1987!Since such seats are bought by professionals, this was another gaugeof negative professional sentiment,

7. Another characteristic of REMs is gloom and-doom in the finan-cial publishing industry, which the following quotation exempli-fies.

No Boom on Wall Street for PrintersWhile the stock market reaches new highs, the Crash of

1987 is still taking its toll on the financial printing business.Sorg Inc., one of the nation’s oldest and largest financial print-ers, filed for protection under Chapter 11 of the Federal Bank-ruptcy Code, joining Charles P. Young, founded in 1902, whichfiled for bankruptcy earlier this month, and Packard Press andits parent company, the Basix Corporation, which filed in Janu-ary 1988. It will not be long, analysts and industry experts say,before these companies are joined by others as the industrycontinues to reel.

The New York Times, 12 Aug 89

Normally, one would think the bull market of that time wouldhave meant prosperity for the financial publishing industry.At this same time, Charles Githler’s Investment Seminars Interna-tional held a seminar in San Francisco. I noted how much smallerthe crowds were than in August 1987, just before the market brokedawn. Since tickets run $700 apiece, attendance is obviously notcasual, and as such it is an important gauge of investor sentiment.Charles Githler told me that he had had a record 750 attendees inAug 87, and a record low of around 350 the following year.In 1989 according to a number of newsletter editors, the industrywas in its worst depression ever. It has been my experience overthe last 35 years that new subscriptions remain very low all the wayup in a bull market, until a Top is approached. By the time thepublic turns bullish enough to subscribe to a newsletter, a Top is

no longer far away.8. Self-observation is a crucial aspect of self control. As a RBM rises

in its final phases, excitement and greed will become palpable.Remember the theory that all gamblers secretly want to lose; it’sthe source of mass masochism that leads the public to sell out atBottoms and to buy at Tops.The same thinking can be applied to the decline in the price of aseat on the New York Stock Exchange, mentioned above. Normally,one would expect such seat prices to rise with the market, becausemore public participation means higher commissions, but in aRBM this Indicator is reversed.

9. As an example of how self-observation becomes crucial during aREM, if you are afraid to buy, if you tremble at the thought that assoon as you buy something the stock market is going to collapse,that is bullish. It is when you take a second mortgage on yourhome to buy stocks, or when you are extremely confident whenbuying, that you should recognize the personal sign to get out.

10.The above nine points are a highly intangible evaluation. RBMstend to move with great rapidity so we have to institute “trigger”points, just in case. Remember, deep Corrections are not a fea-ture of RBMs, so at the first sign of a deep Correction it becomesnecessary to take precautions. How deep is deep? Well, there isno easy answer to that one, and we will all have to use our ownjudgments and limits.

Excerpted from The Dines Letter, September 1, 1989

BIOGRAPHY

James Dines, Editor of The Dines Letter (PO Box 22, Belved-ere, CA 94920) and a well-known gold bug, has written severalbooks on technical analysis including The Invisible Crash andHow the Average Investor Can Use Technical Analysis for StockProfits.


THIS ARTICLE APPEARED IN THE MTA JOURNAL, SUMMER-FALL 1994

In the following article I will discuss the analysis of a Trading Range,employing terms and principles developed by Richard Wyckoff inthe 1920s and 30s and more recently by the “Stock Market Institute.”In technical analysis, there are a variety of methods used to analyzetrading range formations and forecast the expected direction andextent of the move out of a trading range. Most practitioners of tech-nical analysis, whether familiar with the Wyckoff method or not, willbe able to relate many of the points and principles being discussed tothose they are already familiar with.

Much of Wyckoff's analysis and working principles were based onwhat he identified as three fundamental laws:1. The Law of Supply and Demand – which simply states that when

demand is greater than supply, prices will rise, and when supply isgreater than demand, prices will fall.

2. The Law of Cause and Effect – postulates that in order to have aneffect you must first have a cause, and that effect will be in pro-portion to the cause. This law’s operation can be seen working, asthe force of accumulation or distribution within a trading rangeworks itself out in the subsequent move out of that trading range.Point and figure chart counts can be used to measure this causeand project the extent of its effect.

3. The Law of Effort vs. Result – helps us evaluate the relative domi-nance of supply vs. demand, through the divergence or dishar-mony between volume and price, when considering relativestrength, comparative price progress and trading volume.An objective of Wyckoff analysis is to aid in establishing a specula-

tive position in correct anticipation of a coming move where a favor-

ANATOMY OF A TRADING RANGE

Jim Forte

able reward/risk ratio exits (at least 3 to 1) to justify taking that posi-tion. Trading Ranges (TR’s) are places where the previous move hasbeen halted and there is relative equilibrium between supply anddemand. It is here within the TR that dominant and better-informedinterests conduct campaigns of accumulation or distribution in prepa-ration for the coming move. It is this force of accumulation or distri-bution that can be said to build a cause which unfolds in the subse-quent move.

Because of this building of force or cause, and because the priceaction is well defined, trading ranges represent special situations thatoffer trading opportunities with potentially very favorable reward/risk parameters. To be successful however, we must be able to cor-rectly anticipate the direction and magnitude of the coming moveout of the trading range. Fortunately, Wyckoff offers us some guide-lines and models by which we can examine a trading range.

A preview of the guidelines and model schematics presented here, along withthe accompanying explanation of the terms and principles represented in theschematics, will go a long way to further the reader’s understanding of the text.

It is through the identification and analysis of the price and vol-ume action and certain principles in action within the various phasesof the trading range (TR) that the trader can become aware and con-clude that supply or demand is becoming dominant and correctlyanticipate the coming move. It is through the analysis of the phasesof the TR that we can distinguish accumulation/re-accumulation fromdistribution/redistribution.

The Wyckoff method employs bar charts along with certain termsand principles in action to determine the expected direction andtiming of a coming move. It also employs point and figure chart

6


counts to aid in projecting the extent of the move.For those interested in exploring the use of point and figure charts,

references are available from the Wyckoff “Stock Market Institute”(SMI) and from other sources on technical analysis. Our emphasishere will be primarily on the analysis of bar chart formations.

The following illustrations represent an idealized Wyckoff modelof market cycles involving supply and demand, accumulation anddistribution, and a conception of the primary market phases.

ACCUMULATION

Schematic 1 is a basic Wyckoff model for accumulation. While thisbasic model does not offer us schematic for all the possible variationsin the anatomy of the TR, it does provide us a representation of theimportant Wyckoff principles, often evident in an area of accumula-tion, and the identifiable phases used to guide our analysis throughthe TR toward our taking of a speculative position.

Phase AIn Phase A, supply has been dominant and it appears that finally

the exhaustion of supply is becoming evident. This is illustrated inPreliminary Support (PS) and the Selling Climax (SC) where widen-ing spread often climaxes and where heavy volume or panicky sellingby the public is being absorbed by larger professional interests. Onceexhausted an Automatic Rally (AR) ensues and then a SecondaryTest (ST) of the selling climax. This Secondary Test usually involvesless selling than on the SC and with a narrowing of spread and de-creased volume. The lows of the Selling Climax (SC) and the Sec-ondary Test, and the high of the Automatic Rally (AR) initially setthe boundaries of the trading range. Horizontal lines may be drawnhere to help us focus our attention on market behavior in and aroundthese areas.

It is also possible that Phase A can end without dramatic spreadand volume, however it is usually better if it does, in that more dra-matic selling will generally clear out all the sellers and clear the wayfor a more pronounced and sustained markup.

Where a TR represents re-accumulation (a trading range within acontinuing up move), we will not have evidence of PS, a SC, and STas illustrated in phase A of Schematic 1. Phase A will instead lookmore like Phase A of the basic Wyckoff distribution schematic (Sche-matic 2 or 3); but none the less, Phase A still represents the area ofthe stopping of the previous move. The analysis of Phase B throughE would proceed the same as is generally advised within an initialbase area of accumulation.

Phase BIn Phase B, Supply and Demand on a major basis are in equilib-

rium and there is no decisive trend. The clues to the future course ofthe market are usually more mixed and elusive, however here aresome useful generalizations.

In the early stages of Phase B the price swings tend to be ratherwide, and volume is usually greater and more erratic. As the TR un-folds, supply becomes weaker and demand stronger as professionalsare absorbing supply. The closer you get to the end or to leaving theTR, volume tends to diminish. Support and resistance lines, (shownas horizontal lines A, B, C, and Don the Accumulation Schematic 1)usually contain the price action in Phase Band will help define thetesting process that is to come in Phase C. The penetrations or lackof penetrations of the TR enable us to judge the quantity and qualityof supply and demand.

Phase CIn Phase C, the stock goes through a testing process. The stock

may begin to come out of the TR on the upside with higher tops and

bottoms or it may go through a downside spring or shakeout, breakingprevious supports. This latter test is preferred, given that it does abetter job of cleaning out remaining supply from weak holders andcreates a false impression as to the direction of the ultimate move.Our Schematic 1 shows us an example of this latter alternative.

Until this testing process, we cannot be sure the TR is accumula-tion and must wait to take a position until there is sufficient evidencethat mark-up is about to begin. If we have waited and followed theunfolding TR closely, we have arrived at the point where we can bequite confident of the probable upward move. With supply appar-ently exhausted and our danger point pinpointed, our likelihood ofsuccess is good and our reward/risk ratio favorable.

The shakeout at point 8 on our Schematic 1 represents our firstprescribed place to initiate a long position. The secondary test atpoint 10 is better, since a low volume pullback and a specific low riskstop or danger point at point 8 gives us greater evidence and moreconfidence to act. A sign of strength (SOS) here will bring us intoPhase D.

Phase DIf we are correct in our analysis and our timing, what should fol-

low here is a consistent dominance of demand over supply as evi-denced by a pattern of advances (SOSs) on widening spreads andincreasing volume, and reactions (LPSs) on smaller spreads and di-minished volumes. If this pattern does not occur, then we are ad-vised not to add to our position and look to close our original posi-tion until we have more conclusive evidence that markup is begin-ning. If our stock progresses as stated above, then we have additionalopportunities to add to our position.

Our aim here is to initiate a position or add to our position as thestock or commodity is about to leave the trading range. At this point,the force of accumulation has built a good potential and could beprojected by using the Wyckoff point and figure method (or perhapsanother method of the reader’s own choosing).

We have waited to this point to initiate or add to our positions inan effort to increase our likelihood of success and maximize the useof our trading capital. On our Schematic 1, this opportunity comesat point 12 on the “pullback to support” after “jumping resistance”(in Wyckoff terms this is known as “Backing Up to the Edge of theCreek” after “Jumping Across the Creek”). Another similar opportu-nity comes at point 14, a more important point of support and resis-tance.

In Phase D, the mark-up phase blossoms as professionals begin tomove up the stock. It is here that our best opportunities to add toour position exist, before the stock leaves the TR.

Phase EIn Phase E, the stock leaves the TR and demand is in control.

Setbacks are unpronounced and short lived. Having taken our posi-tions, our job here is to monitor the stock’s progress as it works outits force of accumulation. At each of points 8, 10, 12, and 14 we maytake positions and use point and figure counts from these points tocalculate price projections and help us to determine our reward/riskprior to establishing our speculative position. These projections willalso be useful later in helping us target areas for closing or adjustingour position.

Remember our Schematic 1 shows us just one idealized model oranatomy of a trading range encompassing the accumulation process.There are many variations of this accumulation anatomy and we ad-dressed some of these considerations earlier. The presence of a Wy-ckoff principle like a selling climax (SC) doesn’t confirm that accu-mulation is occurring in the TR, but it does strengthen the case forit. However, it may be accumulation, redistribution or nothing. The


Accumulation SchematicPhases A through E: Phases through which the Trading Range

passes as conceptualized by the Wyckoff method and explainedin the text.

Lines A and B... define support of the Trading Range.Lines C and D... define resistance of the Trading Range.(PS) Preliminary Support is where substantial buying begins

to provide pronounced support after a prolonged downmove.Volume and spread widen and provide a signa1 that thedownmove may be approaching its end.

(SC) Selling Climax... the point at which widening spread andselling pressure usua1ly climaxes and heavy or panicky sellingby the public is being absorbed by larger professional interestsat prices near a bottom.

(AR) Automatic Rally... selling pressure has been pretty muchexhausted. A wave of buying can now easily push up priceswhich is further fueled by short covering. The high of thisrally will help define the top of the trading range.

(STs) Secondary Test(s)... revisit the area of the Selling Cli-max to test the supply demand balance at these price levels. Ifa bottom is to be confirmed, significant supply should not re-surface, and volume and price spread should be significantlydiminished as the market approaches support in the area ofthe SC.

The “CREEK” is an ana1ogy to a wavy line of resistance drawnloosely across rally peaks within the trading range. There are

of course minor lines of resistance and more significant onesthat will have to be crossed before the marketís journey cancontinue onward and upward.

Springs or Shakeouts usually occur late within the tradingrange and a1low the market and its dominant players to makea definitive test of available supply before a markup campaignwill unfold. If the amount of supply that surfaces on a break ofsupport is very light (low volume), it will be an indication thatthe way is clear for a sustained advance. Heavy supply here willusually mean a renewed decline. Moderate volume here maymean more testing of support and to proceed with caution.The spring or shakeout also serves the purpose of providingdominant interests with additiona1 supply from weak holdersat low prices.

Jump Across the Creek (JAC) is a continuation of the creekana1ogy of jumping resistance and is a good sign if done ongood spread and volume – a sign of strength (SOS).

Sign of Strength (SOS)... an advance on good (increasing)spread and volume.

Back Up (BU) to a Last Point of Support (LPS) – a pull back tosupport (that was resistance) on diminished spread and vol-ume after a SOS. This is good place to initiate long positionsor to add to profitable ones.

Note: A series of SOSs and LPSs is good evidence that abottom is in place and Price Markup has begun.


use of Wyckoff principles and phases identifies and defines some ofthe key considerations for evaluating most any trading range and helpsus determine whether supply or demand is becoming dominant andwhen the stock appears ready to leave the trading range.

DISTRIBUTION

Accompanying our discussion of distribution are Schematics 2 and3, two variations of the Wyckoff model for distribution. While thesemodels only represent two variations of the many possible variationsin the patterns of a distribution TR, they do provide us with the im-portant Wyckoff principles often evident in the area of distributionand the phases SMI uses to guide our analysis through the TR towardtaking a speculative position.

Much of this discussion and analysis of the principles and phasesof a TR preceding distribution are the inverse of a TR of accumula-tion, in that the roles of supply and demand are reversed.

Here, the force of “jumping the creek” (resistance) is replaced bythe force of “falling through the ice” (support). Given this, I will notrepeat all the points made earlier, but rather emphasize those areaswhere the differences merit discussion and where additional pointsneed to be made or reemphasized. It is useful to remember thatdistribution is generally accomplished in a shorter time period ascompared to accumulation.

Phase AIn Phase A, demand has been dominant and the first significant

evidence of demand becoming exhausted comes at point 1 at Pre-liminary Supply (PSY) and at point 2 at the Buying Climax (BC).(See Schematic 2 and 3.) It often occurs on widespread and climaticvolume. This is usually followed by an Automatic Reaction (AR) andthen a Secondary Test (ST) of the BC, usually on diminished vol-ume. This is essentially the inverse of Phase A in accumulation.

As with accumulation, Phase A in distribution may also end with-out climactic action and simply evidence exhaustion of demand withdiminishing spread and volume.

Where Redistribution is concerned (a TR within a larger continu-ing downmove), we will see the stopping of a downmove with or with-out climactic action in Phase A. However, in the remainder of theTR the guiding principles and analysis within Phases B through Ewill be the same as within a TR of a Distribution market top.

Phase BThe points to be made here about Phase Bare the same as those

made for Phase B within Accumulation, except clues may begin tosurface here of the supply I demand balance moving toward supplyinstead of demand.

Phase COne of the ways Phase C reveals itself after the standoff in Phase B

is by the “sign of weakness” (SOW) shown at point 10 on Schematic2. This SOW is usually accompanied by significantly increased spreadand volume to the downside that seems to break the standoff in PhaseB. The SOW mayor may not fall through the Ice,” but the subse-quent rally back to point 11, a “last point of supply” (LPSY) is usuallyunconvincing and is likely on less spread and/or volume.

Point 11 on both Distribution Schematics 2 and 3 give us our lastopportunity to cover any remaining longs and our first inviting op-portunity to take a short position. Even a better place would be onthe rally testing point 11, because it may give us more evidence (di-minished spread and volume) and/or a more tightly defined dangerpoint.

Looking now at Schematic 3, Phase C may also reveal itself by apronounced move upward, breaking through the highs of the TR.

This is shown at point 11 as an “Upthrust After Distribution” (UTAD).Like the terminal shake out discussed in accumulation, this gives afalse impression of the direction of the market and allows furtherdistribution at high prices to new buyers. It also results in weak hold-ers of short positions surrendering their positions to stronger playersjust before the downmove begins. Should the move to new highground be on increasing volume and “relative narrowing spread” andthen return to the average level of closes of the TR, this would indi-cate lack of solid demand and confirm that the breakout to the up-side did not indicate a TR of accumulation, but rather a formation ofdistribution.

A third variation not shown here in schematic form would be anup thrust above the highs of the trading range with a quick fall backinto the middle of the TR, but where the TR did not fully representdistribution. In this case, the TR would likely be too wide to fullyrepresent distribution and there would be a lack of concentrated sell-ing except in the latter portions of the TR.

Phase DPhase D, arrives and reveals itself after the tests in Phase C show

us the last gasps or the last hurrah of demand. In Phase D, the evi-dence of supply becoming dominate increases either with a breakthrough the “ICE” or with a further SOW into the TR after an upthrust.

In Phase D, we are also given more evidence of the probable di-rection of the market and the opportunity to take our first or addi-tional short positions. Our best opportunities are at points 13, 15,and 17 as represented on our Schematics 2 and 3. These rallies rep-resent “Last points of Supply” (LPSY) before a markdown cycle be-gins. Our “averaging in” of the set of positions taken within Phases Cand D as described above represent a calculated approach to protectcapital and maximize profit. It is important that additional shortpositions be added or pyramided only if our initial positions are inprofit.

Phase EIn Phase E, the stock or commodity leaves the TR and supply is in

control. Rallies are usually feeble. Having taken our positions, ourjob here is to monitor the stock’s progress as it works out its force ofdistribution.

Successful understanding and analysis of a trading range enablestraders to identify special trading opportunities with potentially veryfavorable reward/risk parameters. When analyzing a TR, we are firstseeking to uncover what the law of supply and demand is revealing tous. However, when individual movements, rallies or reactions arenot revealing with respect to supply and demand, it is important toremember the law of “effort versus result.” By comparing rallies andreactions within the trading range to each other in terms of spread,volume, velocity and price, additional clues may be given as to thestock’s strength, position and probable course.

It will also be useful to employ the law of “cause and """ Withinthe dynamics of a TR, the force of accumulation or distribution givesus the cause and the potential opportunity for substantial tradingprofits. It will also give us the ability, with the use of point and figurecharts, to project the extent of the eventual move out of the TR andhelp us to determine if those trading opportunities favorably meetor exceed our reward/risk parameters.

REAL WORLD EXAMPLES

In addition to the model schematics provided here, some empiri-cal examples of real world trading ranges are also presented (seepages 30-31), where Accumulation/Re-accumulation preceded a


Distribution SchematicsSchematics 2 and 3 show us two model variations of a distri-

bution Trading Range.Phases A through E... phases through which the Trading

Range (TR) passes as conceptualized by the Wyckoff methodand explained in the text.

(PSY) Preliminary Supply... is where substantial selling beginsto provide pronounced resistance after an up move. Volumeand spread widen and provide a signal that the up move maybe approaching its end.

(BC) Buying Climax... is the point at which widening spreadand the force of buying climaxes, and heavy or urgent buyingby the public is being filled by larger professional interests atprices near a top.

(AR) Automatic Reaction... with buying pretty much exhaustedand heavy supply continuing. an AR follows the BC. The low ofthis sell-off will help define the bottom of the Trading Range(TR).

(ST) Secondary Test(s)... revisit the area of the Buying Climaxto test the demand/ supply balance at these price levels. If atop is to be confirmed, supply will outweigh demand and vol-ume and spread should be diminished as the market ap-proaches the resistance area of the BC.

(SOW) Sign of Weakness... at point 10 will usually occur onincreased spread and volume as compared to the rally to point9. Supply is showing dominance. Our first “fall on the ICE”holds and we get up try to forge ahead.

The ICE... is an analogy to a wavy line of support drawnloosely under reaction lows of the Trading Range. A break

through the ICE will likely be followed by attempts to get backabove it. A failure to get back above firm support may mean a“drowning” for the market.

(LPSY) Last Point of Supply... (Schematic 2/Point 11): afterwe test the ICE (support) on a SOW, a feeble rally attempt onnarrow spread shows us the difficulty the market is having inmaking a further rise. Volume may be light or heavy, showingweak demand or substantial supply. It is at these LPSY’s thatthe last waves of distribution are being unloaded before mark-down is to begin.

Schematic 2/Point 13: after a break through the ICE, a rallyattempt is thwarted at the ICE’s surface (now resistance). Therally meets a last wave of supply before markdown ensues.

LPSY's are good places to initiate a short position or to addto already profitable ones.

(UTAD) UPthrust After Distribution... (See Schematic 3/ Point11). Similar to the Spring and Terminal Shakeout in the tradingrange of Accumulation, a UTAD may occur in a TR of distribu-tion. It is a more definitive test of new demand after a breakoutabove the resistance line of the TR, and usually occurs in thelatter stages of the TR.

If this breakout occurs on light volume with no followthrough or on heavy volume with a breakdown back into thecenter of the trading range, then this is more evidence thatthe TR was Distribution not Accumulation.

This UTAD usually results in weak holders of short posi-tions giving them up to more dominant interests, and also inmore distribution to new, less informed buyers before a sig-nificant decline ensues.


Long Term Accumulation

Phase A: Shows us the PS & SC with the exhaustion of supply as the steep downtrend inending. The AR & ST set the approximated boundaries of the TR to follow.

Phase B: In the early stage, we see a wide swing & higher vol, and the first signs of demandasserting its dominance, as professionals are absorbing supply. Late in Phase B, lowvol shows supply has dwindled at the TR lows.

Phase C: Gives us a final unconvincing test and break of the TR lows on extremely lightvolume. This is followed by a SOS on dramatically increased volume.

Phase D: We see a consistent & pronounced dominance of demand over supply on widen-ing spreads and increased volume to the upside. Reactions are comparatively weakand on light volume.

Phase E: The stock is marking up on rising volume. Demands remains in control.

Markup, and Distribution preceded a Markdown. While these em-pirical examples may not fit the idealized schematics exactly, I haveidentified and annotated on each of the chart examples, the Wyckoffprinciples in action and the five Wyckoff phases of a trading range.

BIBLIOGRAPHY

1. Hutson, J., Weis, D., and Schroeder, C., Charting the Market, TheWyckoff Method, Technical Analysis of Stocks and Commodities, Se-attle, 1990.

2. Pruden, H.O. and Fraser. B., The Wyckoff Seminars, Golden GateUniversity, San Francisco, Fall 1992 and Spring 1993.

3. Wyckoff/Stock Market Institute, literature, illustrations, and au-dio tapes. 13601 N. 19th Avenue, Suite 1, Phoenix, Arizona 85029,Tel: 602/942.5581, Fax: 602/942.5165.

4. Charts supplied by Telescan 3.0, Houston. Texas.

BIOGRAPHY

Jim Forte has been using technical analysis professionally andpersonally in both stocks and commodities since 1986. He iscurrently employed in the research services department of amajor brokerage firm where he maintains a market update ser-vice. He studies and teaches Technical Analysis at Golden GateUniversity in San Francisco and also offers seminars. He is aprofessional member of IFTA and the TSAA of San Francisco.


Intermediate Reaccumulation

Phase A: Stops previous move.Phase B & C: Shows comparatively weak volume on consolidation as stock moves down.

Volume very light on series of lower lows on Shakeouts. No new supply on #3 spring.Demand showing dominance as stock comes off spring.

Phase D: Shows continuing pattern of demand in control. Gives us sufficient evidence toadd to our longs on pullback

Phase E: Stock Marking Up. Demands in control.

Intermediate Reaccumulation

Phase A: Shows Buying Climax stopping previous up move and more pronounced prelimi-nary support and selling climax facilitating accumulation into stronger hands.

Phase B: Inconclusive evidence but does show us evidence of rally on good spread andvolume.

Phase C: Shows final low on diminished volume compared to ST and holds support areaabove climax low. Move off of low shows pattern on expanding spread and volume.

Phase D & E: Continues pattern of Demand in Control.

Distribution

Phase A: Shows us PSY and Push to new highs (BC) on falling volume. ST fails and closesbelow BC high. The subsequent downward immediately precedes. The next attempt,a few days later, is on poor volume and cannot reach previous light.

Phase B: Gives us some early clues that supply is in control. Bearish activity is evidentshowing a SOW on increased volume and the rallies on comparatively low volumeindicating a lack of demand. Phase B also shows a breakthrough the TR Support Lines.Subsequent rallies are also on poor volume. Additional Breaks of Support line oneven higher volume.

Phase C: We break through the ice and manage to rise above it, however, volume in uncon-vincing. We can only rise to meet resistance at the supply line and the bottom of ourinitial trading range. This gives us as LPSY and an opportunity to take a short positionwith a well-delineated risk just about the previous high at 19 1/2.

Phase D: We fall through the ice again, but on significantly higher volume. We have norallying power and a feeble attempt to reach the ice fails. Supply has continued itsdominance. We are given a last opportunity to add to our short position on the rallyback to the ice.

Phase E: Markdown accelerates and supply is in control.

Distribution

Phase A: We see the up-move stopped by PSY and the BC. We have and AR and an ST.Phase B: In phase B relative equilibrium on low volume. No clear indications seem re-

vealed by a #3 spring before the upthrust.Phase C: As in our #3 Schematic, MTA however shows us a UTAD and then quickly returns

to the trading range. The UTAD follows the right side of the TR in phase C.Phase D: Shows a progression of declines and rallies with higher volume on the down

swings. A Supply Line is evident. Breaks through the ice.Phase E: Our rally back to the ice fails and markdown accelerates.


The purpose of this review paper is to list, explain, and evaluate severalwell-known stock market sentiment indicators over many periods of time. Theseindicators include Option put/call ratios, advisory letters, short interest, mu-tual fund cash, and other contrary, against-the-crowd statistics.

The reason that this is a Review article rather than Research is that therehas been much written on these indicators by the experts of the industry (al-though very little recently, which I hope to update). Each indicator’s peaksand troughs will be juxtaposed with the appropriate index or average. I in-tend to first define and describe each Indicator and assess its efficacy; then, ina Discussion section. I place each on a Bell/Growth Curve model in its appro-priate place in time.

These findings should be of use to anyone who needs to ascertain marketdirection and reversals for trading.

Much has been written over the years about contrary opinion; ithas become widely accepted and clever to go against the crowd –“When everyone looks one way, look the other!” Although primarilya true concept, there are a few considerations I would like to bring tolight. Most serious investors are familiar with the South Seas Bubbleand Tulip Bulb Mania from Mackay’s Extraordinary Popular Delu-sions and the Madness of Crowds. Although history repeats itself, italmost never does it exactly in the same way. Try developing a tulipcraze in Holland today, or, observe the Deutschbank’s tight stanceagainst inflation after the wheelbarrows of DMarks decades ago.

In his talk at the 1994 annual TSAA conference in San Francisco,John Bollingcr stressed how important it is to know against whom tobe contrary. Should one take a position against the world being round,or the sun rising tomorrow? Rather, the successful investor has toestablish, through introspection, an internal monitor which will warnhim when he has stopped doing his own analysis and has begun rely-ing on peers, media items, or a guru for opinions, “tips,” and timing.

In his book, Humphrey Neill explains that contrary opinion isnot necessarily cynical or negative, but sees both sides of an issueusing one’s experience and logic to see reality. Just as some oscilla-tors can be useful in the middle of a trend but wrong at the extremes,so are the majority often correct during a Bull or Bear market butmanically wrong when it reverses, especially when they are requiredto act, like buy or sell, rather than just observe. Examples of herdlogic at these junctures are “This time it is different,” or “What canpossibly go wrong.” or at the nadir, “This company is doing every-thing wrong – it’s hopeless.” In the following pages I would like toillustrate which indicators are the most effective in forecasting mar-kets, individually and in combination.

One category of sentiment measurement is the surveys found inBarron’s and elsewhere on advisors, letter writers and investors. Al-though the majority or these surveys only go back a few years, theirroots can be found (according to Neill) in an SEC poll before theCrash of ’46 where advice from Brokers and Advisors showed a bear-ish percent of only 4.1%. Earl Hadady of the Bullish Consensus feelsso strongly about this indicator that he feels (in his excellent articlein the 1986 MTA Journal) that Polling is a third and most importantmethod of analysis, above Technical and Fundamental. The basicquestion of why investors bought or sold (the public needs answers,the media attempts to fill that need, either in honest attempts or in

ANSWERING THE BELL OF SENTIMENT INDICATORS

Brent L. Leonard

some cases intentionally misleading) is not important; rather whatthe public is really doing, as manifest in the Technical signals of Priceand Volume over Time. Unfortunately, just as the media and econo-mists range widely in their beliefs and advice, so do technically-ori-ented gurus and letter writers. As Hadady points out, extreme ex-amples (70% or more) occur less than once a year. If 80% are of onemind, only 1 of 5 traders (especially in zero-sum Futures markets)hold a contra position – therefore they are the strong hands of Rich-ard Wyckoff’s Composite Operator, or the Big Money that controlsmarkets), impervious to margin calls or scared money and in no hurryto get out without a large profit when the majority is sated – as indi-cated when favorable news now has no effect. It is at this point thatshorts are covered, margins are full, and complacency is rampant.

In summation then, by way of paraphrasing into an anagram,Edwin Lefevre in Reminiscences of a Stock Operator, the mottoF.I.G.H.T. could represent Fear, Ignorance, Greed, and Hope overTime exemplifying the emotions which we need to control to be theultimate, dispassionate Composite Operator, or ideal trader.

One way to analyze markets by the notion that there is a “control-ling factor” or Wyckoffian Composite Operator behind market move-ments was portrayed in a white paper written by Dr. Henry “Hank”Pruden for a class at Golden Gate University. He likens the marketto a clothing Fashion Cycle wherein one or more top designers in thehaute couture world decides a new dress length, style, color is needed,it is then created and diffused throughout the fashion elite, adoptedand imitated by the general public, until the last housewife in a farmcommunity in the Midwest has given in to the new look. Magazines,stores, media shows have “told” the public what to wear, driving exist-ing dresses, ties, and other clothing into premature obsolescence.Indeed, if print and television media can “hype” or market athleticevents, songs and movies, why not glamour stocks, mutual funds andother securities?

THE INDICATORS

The odd-lot short ratio is derived by taking odd-lot purchases addedto odd-lot sales, dividing by two (much like open interest in Futuresis obtained), and dividing that into odd-lot short sales. I did not findthis indicator an effective contrary tool, especially in relation to itssuccess before the current bull market, for the following reasons:only 2 major spikes above 15 occurred in this 12-year time frame (seechart 1). Although both proceeded large up moves, they were theresult of a sideways trading range (1986) and a sharp sell off (1990).However, seven other smaller spikes above 10 did not render bullmarkets. Conversely, low readings did not indicate down moves inthe market with three exceptions – 1987, 1990, and mid-1991 – ver-sus several that preceded up moves. Other reasons might includethese: smart money was shorting in sma1l odd-lots to avoid the uptickrule, now extant in over-the-counter stocks; some shorting was usedin a derivative fashion to hedge and box positions, more than in thepast; many odd-lotters with scarce money moved to index and equityoptions over the past fifteen to twenty years.

7THIS ARTICLE APPEARED IN THE MTA JOURNAL, SPRING-SUMMER 1996


NYSE Short Ratio; S&P 500Merrill Lynch data 1965-94

Looking at monthly data on NYSE short interest ratio and its ef-fect on the S&P 500 Index, historically this was an accurate measureof contrary opinion, where the early adopters of trend were correctand profitable, and those at the manic end (see arrows on the leftside of the bottom part of Chart 2) were 180% wrong. Sharp rallies,abetted by short covering, ensued in cyclic fashion. Once we ended

the 17-18-year trading range cycle and started the current bull mar-ket of 1982, things noticeably changed: shorting became and re-mained excessive, again mostly due to derivative hedging whereinshorts do not have to be covered and strong hands do not have tomeet margin ca1ls. Another factor to consider is that currently over10% of the NYSE is Closed End funds, mostly bond and country types.Sti1l, as the arrows continue to show, rising spikes seem to jibe withup moves on the S&P 500, with the one exception.

Specialist Short Sales Vs. PublicMerrill Lynch data 1978-94

What appears to be a better indicator of shorting sentiment, al-though far from perfect, is the Specialist versus Public ratio, shownbelow (Chart 3). Specialists are the closest persons to buyers’ andsellers’ decisions, although there is a one- to two-week delay in find-ing their actions. We can observe that not only are the Buy and Sellsignals mostly accurate (B & S not mine), with an occasional misfire(0), but over the long haul, timing market trades would afford youbetter than 50% gain over buy-and-hold. The “middle clip” in Charts4 & 5 refers to the areas between the dotted lines, lower half.

Mutual Fund Cash RatioICI – Ned Davis Research 1978-93

Chart 4 illustrates how excessive cash can powermarkets upward while, at least in a major Bull mar-ket, too little doesn’t always correlate to a major de-cline. One reason for this is that the pressure of short-term performance, especially with “Money Manage-ment Consultants” demanding low cash ratios for cli-ents, poses the threat of moving them to anothermoney manager who will “rotate” the cash into an-other sector.

In addition to the fact that excess mutual fundcash does precede rallies, the reciprocal occurrenceof mutual fund buying climax (as depicted in the NedDavis chart 5) precedes either substantial declines,or at least long, sideways trading ranges. Inversely,from the 1987 Crash until well into 1989, Mutual Fundredemptions exceeded sales throughout that up mar-ket, just in time to buy (A) into the next decline (B).

Margin Debt 1967-93Merrill Lynch data

As the long-term chart indicates (Chart 6 with theopposing arrows) Margin Debt has historically beena correct indicator of major tops, especially in 1973,just before 1982 and dramatically in 1987. After the1990 correction caused the last Margin debt recon-ciliation or covering, the chart shows a straight uptrend, reflecting the investing consumer's,government’s and even global appetite for spendingon credit. Although accurate, like many oscillatorsthe trend can stay in its extreme mode seemingly in-definitely only warning of its imminent bursting.

As I mentioned earlier in the odd-lot short para-graph, when the option market got popular, especially in March 1983with the advent of the OEX (S&P 100 Index), the least accurate oftraders, the under-financed public, switched from odd lots of stockto options on stocks and indices. At the present time, more than1500 stocks, or 75% of the stock market capitalization, have equityoptions. The number of sector indices has also burgeoned dramati-cally. It has been commonly thought that when put volume heavilyoutnumbers call volume, this is a contrary indicator that the marketor underlying entity will rise. This is true for the short-term day trader;

Chart 1

Chart 2

Chart 3


however, looking at the history of the OEX on a weekly basis (Chart7), the opposite seems to be true. Over the 12 years, using Reuter'sparameters of below 0.75 OEX put/call ratio as bearish and over 1.50being bullish, we can see the high numbers are almost always at thetop, providing the put buyers correct. Similarly, the lower numbersconsistently occur at or near the bottoms, when the call buyers wouldbenefit, especially in the mid-1985-86 span.

Curiously, from August 17, 1987 to October 16, 1987 the OEXput/call ratio was locked in a 60-100 range, actually rising into thelast few days before the crash (theoretically bullish). The highestreading ever was in late 1983 – 9.28 – interestingly, just before the bigdecline of January 1984 of some 30 OEX points.

Being a veteran Option Specialist for the OEX’s largest tradingfirm and author of an article in the 1993 MTA Journal (#41) onV.O.I.C.E., a treatment of OEX Volume, and Open Interest input intoa TRIN formula with excellent results (as did Jim Martin, Ray Hines,

John Bollinger, and others in slightly different ways),I was quite surprised by these findings. Obviously fur-ther study using moving averages, and daily data ab-stracts are necessary to verify this conundrum. Look-ing at current daily data in the next chart, we do see amore positive correlation between high put volume,both in the OEX and all-equity CBOE charts, and up-ward price movement. This is fine for day and shortterm trading, but I cannot use a high coefficient inmy Intermediate, positive-trading Master Indicator, forwhich I am currently collecting data and fine-tuning,possibly for a future paper.

VIX IndexCBOE 1983-91

Just a brief word about the VIX index, which mea-sures the Volatility of the OEX index (S&P 100) frommid 1985 on, as shown in Chart 8 above. It actuallydepicts the implied volatility of 8 OEX options, in andout the money, near months. Since it has only beenaround in a Bull market, its only consistent behaviorseems to be: down or nonvolatile during moves of themajor uptrend, with sharp up spike when the marketdeclines, and flat or coiling during trading ranges.Chart II shows the historical high of 150 in the Crashof 1987, and single digit lows during 1993 and 1994,possibly an harbinger of things to come. Although theVIX is very good for trading strategies (buying or sell-ing options depending on the volatility), I find it lessuseful than the Option Premium Ratio, which com-bines put/call sentiment with volatility (see next page).

Option Premium Ratioby Christopher Cadbury, 1986-94

Stocks & Commodities MagazineA rather recent indicator that has established many

valid instances, primarily due to extensive research andseveral articles by Christopher Cadbury (to whom I owemuch gratitude for endless data), is the Option Pre-mium Ratio (OPR). This can only be found in theSentiment Window, Chart Page of Investor’s BusinessDaily, item #5, and essentially combines Put/Call Op-tion sentiment with Implied Volatility of the VIX, onlyit includes all equity options, not just the OEX index.Based on data from 10 years, (although listed Optionshave been around over twenty) dividing put premiumsby call premiums has ranged from .03 to a high of 1.74.

Cadbury established that values below .29 and above 1.18 indicate acontinuation of the trends down and up respectively – like extremelevels of other oscillators. Conversely, OPR’s from 30 to mid-60s gen-erate buy signals and levels to 1.18, sell signals in about 200 differentcombinations of occurrences.

Most of these abstracts are proven almost unanimously by 10 to 20test examples, such as, “Four consecutive day’s of gains or unchangedvalues for the OPR starting from .32 to .51 have always producedsignificant rallies in the stock market. A few, however, such as “Iden-tical values for the OPR in the range between .80 to .88 separated by5 to 7 days have always produced significant declines in the stockmarket have insufficient testing and border on the “whenever I weara red tie on Friday the market goes up for 3 days” category. Table 1 isa data table of one of the most heavily tested “pattern recognition”examples: it includes the date the 5-7 day series began and the OPRvalues; the next four columns list the number or Dow Jones points

Chart 4

Chart 5


and days just before the event, and the number of points in the sub-sequent rally with the number of days or weeks to complete it. Morewill be heard from on this excellent indicator – I intend to include itin my Master Indicator.

Sentiment Indicators – OpinionsBarron's Polling Surveys

The following section discusses the derivation of the 4 major Sen-timent surveys from newsletters along with charts which show Buyand Sell signals and their respective effectiveness, as shown by %Gains- again this paper is to review, not to research the gathering details.Most effective, I found, were the Market Vane and AAII Newsletters.1. Investor’s Intelligence 1966-95. Investor's Intelligence is published

by Michael Burke’s Chartcraft, and expresses the opinions of over100 advisory letters every week on CNBC and later in Barron's.Since 1966, this has been an excellent contrary indicator with its“trading range” giving its best signals from high 30s (% of Bulls)as a Buy signal and mid-70s as a Sell. Although the Buy signalshave proven very consistent, the Sell indications, which before1989 were quite consistent although very early (sometimes sev-eral months), have been effective in signaling trading ranges asour strong market ensues.

2. Consensus, Inc., 1984-94, Kansas City, MO. The Bullish Consen-sus, from Consensus, Inc. in Kansas City, MO, also uses opinionsfrom advisory services, mostly investment advisors from major bro-kerages using house organs versus newsletters. These figures alsoappear on a 900 line and Barron’s on Saturday. As Chart 12 shows,there were a few very minor price reversals on major Sell signals,

especially in the coiling action of both the S&P 500 and the indi-cator the last 3 years. Still, profits would have bested the marketas measured by the Buy-Hold strategy (see upper left corner ofchart). As I write this paper, this indicator has reached a four-yearhigh of 67 (twice), versus a 71 in the first quarter of 1991.

3. Market Vane Corp., 1980-94, Pasadena, CA. An even better senti-ment indicator is found in the Market Vane of Market Vane Corp.,Pasadena, CA. Comprised of 100 of the top Investment Advisorsfrom Brokers, and obtained on Monday each week, informationappears on a 900 phone number and in Barron’s on Saturday ofthat week. Chart 12 indicates a more precise correlation betweenreversals, although again the sell signals in a strong Bull markettend to be more of a re-accumulation trading range than SAR(stop and reverse). Once more, the last several years resemble

Chart 6 Chart 7


coiling action (extension waves of lesser degrees) with lower highsand higher lows in the Indicator. The chart ends with the springof 1994 correction as the O/P line portends a large up move inthe near future. During the writing of this paper, it rallied up to62 for the first time since 1987. At this time, March 25, 1995, it iscuriously near midrange, or 47 to 53 area, not forecasting theselloffs of the previous 3 indicators.

4. American Association of Individual Investors Survey -1987-95. Thefinal Indicator of the Barron's group is the AAII, or American As-sociation of Individual Investors of Chicago, IL, the true retailtrade. With 25 postcards mailed out each day of each week, nearly100 come back with each investor’s opinion of the market for thenext six months. As might be expected, this indicator has an al-

most perfect correlation exemplifying the aforementioned“crowd” syndrome. Gains Per Annum show more than 3 to 1improvement over buy and hold.

DISCUSSION SECTION

In assembling and analyzing all of the above data, whatbecomes increasingly evident is the difference in the timefactor of each. After working nearly a year on constructing aMaster Indicator from the most successful of these SentimentIndicators, it is very apparent that each of them has a differ-ent time frame. For example, the timing of the Put/Call OEXratio is much more short-term than Margin Debt or MutualFund Cash. Not only that, the optimum position on the Bell/Growth Curve (taken from the work of Everett in 1970) onthe next page is quite different. It is only through a corrobo-rating “nesting” of several indicators that we can hope to vali-date the Master Indicator, which would be a great topic for afuture paper. Using Table 2 as a guide, with help from databy Yale Hirsch in his book Don't SeIl Stocks Monday, I will tryto place each Indicator on the Curve on Chart 14 somewherebetween A and E. The graph is a model illustrating a homo-geneous population of investors and sentiment indicators, andnot an actual frequency distribution. The Growth line repre-sents a Price line and an accumulation of the aggregate Indi-cators, while the Bell Curve depicts Volume as well as the tim-ing phases. Beneath the Bell and Growth Curves I have listedthe indicators under study.

Odd-lot shorting would be the highest early in A, withPublic entering in the C segment – they would have to coverby C, with the Specialists starting to short at E.

Mutual Fund Cash would be large at A, fueling the runthrough D, when it would drop into the single digit percent-age. Conversely, Margin would be at its low at A, becomingmanic at C and D, where the rising slope is sharpest. After an

intensive study of the history of the OEX Index, I can only find ituseful in a contrary way on a very short-term basis. Another look atChart 7 shows that in almost all cases, except in tops of 1986 and1987, high numbers were found at tops, low at the bottoms, meaningtraders were correct in the long view. I must say that our current Bullmarket has had high numbers from hedging and from those specula-tors trying to call the top of this market. Similarly, the VIX Indexand the Option Premium Ratio, derived from option premiums ratherthan volume, are short-term, and would therefore be difficult to placeon the Chart.

Finally, Bearish Sentiment and gloom from investment letters andmedia (magazine covers, financial newspapers and TV) respectively,would be persuasive coming into A; they would gradually mutate intocomplacency through C, and outright euphoria and certainly by E.

CONCLUSION

In conclusion, what I have learned in researching and writing thispaper is that although the basic concepts of Sentiment and all ofTechnical Analysis are eternal, some things do change as marketschange. For example, sentiment indicators such as Odd-Lot Short-ing were rendered less effective by other inexpensive derivatives, suchas options.

Also, just as some Oscillators change parameters in Bull versusBear markets, Sentiment indicators are less reliable in cases like thepresent, where the stock market does virtually nothing but rise, withan occasional sideways trading range. Nonetheless, the most effec-tive of the previously reviewed categories, newsletter polling results,

Chart 8

Chart 9


mutual fund cash, specialist short selling, and even option put/callratios, should be monitored for giving reversal signals at extremeexcesses, in conjunction with other technical tools such as cycles, os-cillators, and support/resistance.

Sentiment is as important as any other technical tools used byTechnical Analysts, and will continue to be so as we enter the area of“Behavior Finance” employing Neural Networks to quantify the Psy-chology of Investing.

BIBLIOGRAPHY

• The Crowd by Gustave LeBon, 1982. Cherokee Publishing Co.• The Art of Contrary Thinking by Humphrey B. Neill, 1992, Caxton

Printers• Reminiscences of a Stock Operator by Edwin Lefevre, 1923, Doran,

Fraser Publishers• Don’t SeIl Stocks on Monday by Yale Hirsch, 1986. Facts On File

Publications• MetaStock Technician Odd-lot, 1982-94• Trendlines Odd-lot Short Sales, 1991-95

• NYSE and Specialist Short Sales, Merrill Lynch Data• Investment Company Institute - Mutual Funds• Ned Davis Mutual Fund Buying• VIX Chart, CBOE (Chicago Board of Options Exchange)• Option Premium Ratio by Christopher Cadbury• Merrill Lynch charts on Investor's Intelligence, Consensus, Inc.,

Market Vane Corp., and American Association of Individual In-vestors

• OEX put/call ratio data, Bloomberg News• OEX charts -Reuters/Quotron Advantage AE

BIOGRAPHY

Brent L. Leonard is an Options Specialist at Schwab 500 inSan Francisco, is Vice President of the Technical Securities Ana-lysts Association of San Francisco, and is completing his Master’sin Finance and Level 3 of the CMT designation.

Brent has taught classes in tcchnical analysis at Golden GateUniversity and Schwab University and has lectured be-fore various groups such as A.A.I.I. He has written sev-eral articles on technical analysis both locally and na-tionally.

Brent attended Stanford University and Universityof the Pacific, receiving a degree in education, latercompleting a business curriculum with honors at MesaCollege in San Diego.

Chart 10


Chart 11

Chart 12


Chart 13

Chart 14


COMBINING TECHNICAL ANALYSIS WITH FUNDAMENTAL VALUATIONTO CREATE A RISK INDICATOR FOR THE STOCK MARKET

Jurrien H. Timmer, CMT

THIS ARTICLE APPEARED IN THE MTA JOURNAL, WINTER-SPRING 1998

8ABSTRACT

It is a widely held assumption among stock market professionals that the twoprincipal fundamental drivers of stock market performance over the inter-mediate term are earnings and interest rates, the combination of which formthe dividend discount model. The goal of this paper is to create such anindicator for the S&P 500, and, through traditional technical analysis,develop a trend/momentum-based timing indicator that signals periods ofintermediate term risk. This timing indicator is intended for investmentmanagers for hedging purposes.

The idea of developing the indicator described in this report was inspired byPaul Macrae Montgomery’s presentation at the 1996 MTA Seminar.

PART I: THE FUNDAMENTAL INPUTS

The oldest and most traditional method of "fundamental" valua-tion in the stock market is the dividend discount model. This modeltakes actual or expected earnings or dividends (D) and divides thatnumber by a discount rate (I) in the following formula to arrive at a"fair value" for stocks (P):

P = D/[1+I/100]The numerator and denominator, earnings and interest rates, com-

prise the principal inputs for this fundamentally-driven valuationmodel. For the technical analyst, it may be useful to combine thiskind of fundamental valuation with traditional technical analysis inthe form of trend and momentum studies in order to identify specifictime periods during which the stock market is at risk on the basis ofearnings and interest rates.

In doing so, we first have to chose the fundamental inputs to beused for our study. The two primary options in terms of the numera-tor are earnings and dividends. Both are a manifestation of the sameunderlying fundamental condition of a stock or stock index, but earn-ings tend to fluctuate a bit more than dividends (because the latterare set quarterly by companies). As a result, earnings are probably abetter gauge for valuation purposes, as long as they are taken overmore than one quarter. The next decision is whether to use actualearnings or expected earnings. Actual earnings are conventionallylooked at on a quarterly or four-quarter trailing basis (in order tosmooth out quarter-to-quarter fluctuations), making it a backward-looking or lagging indicator. Since expected earnings are by defini-tion forward-looking (making them a leading indicator), they offer abetter guide for valuation purposes as long as the forecasts are reliable.Reliable in this case means reaching a critical mass in terms of earn-ings estimates by taking the consensus of major research analysts.Since the early 1980's, two firms have been providing consensus earn-ings expectations for the S&P 500 by polling the estimates of majorWall Street firms for the earnings of the S&P 500 on a 12-month-forward basis. The companies are I/B/E/S and First Call. For thispaper, the data from I/B/E/S are used. Chart 1 shows both the lag-ging and leading earnings figures for the S&P 500. The top clip de-picts a weekly bar chart of the stock index, going back to 1982. Themiddle clip shows the expected earnings on a 12-month-forward ba-sis, and the bottom clip shows actual quarterly earnings. While Chart1 shows both estimated earnings and actual earnings, only expected

earnings will used from here on in order to make it a true leadingindicator. Because the earnings estimates begin in 1982, that will bethe beginning of the study.

Chart 1

The next step is to look at the denominator: interest rates. Becauseequities are long-term assets, a long-term interest rate should be used,such as the 30-year Treasury or Moody's long-term BAA corporatebond yield. For the purpose of this exercise, the Treasury long bondis used because data are widely available and because the bond isalways about 30 years to maturity (whereas the Moody's yield reflectsan index which changes over time, making it unclear whether theduration has remained stable over the past 15 years). Now that theinterest rate vehicle has been established, we have to determine howwe are going to discount the earnings numbers. The conventional(orthodox) method divides the earnings number by [1+I/100]. An-other way, however, was recently demonstrated by Paul Macrae Mont-gomery at the 1996 MTA Seminar, and consists of simply dividing theearnings number by [I/100]. We’ll call this the unorthodox method.Chart 2 shows both measures. The middle clip shows the orthodoxmethod (right scale) while the bottom clip shows the unorthodox(left scale). The discount factors have been inverted to show theircorrelation to stock prices.


Chart 2

Note that, while both series look exactly the same, the bottomnumber is much larger (e.g. 14.14 vs. 0.934 as of September 13, 1996).As a result, when both series are multiplied by the numerator (orwhen the inverse is divided), the latter approach will cause a muchbigger impact on the product of the discounting equation. The ef-fect of both methods is shown on Chart 3.

Chart 3

The top clip shows again the weekly S&P 500. The middle clipshows expected earnings using the orthodox approach, and the bot-tom clip shows the effects of the unorthodox approach. It becomesimmediately apparent that the second discounting method reflects

the course of interest rates in a much more pronounced way. Thusrather than just reflecting the earnings outlook, this series now trulycombines the effect of both earnings expectations and the course ofinterest rates. Because we want to create an indicator that uses bothof these drivers, the unorthodox discounting formula of D/[I/100]is the one we will use to build our technical indicator.

Now that we have created the time series (which will be called "E/I" from now on) on which to build our trend/momentum indicator,it is useful to show what the correlation actually is between our com-puted study and the S&P 500. Chart 4 shows both time series on alog scale.

Chart 4

The chart nicely illustrates how the stock market usually corre-lates with E/I, but that there are times when significant bearish diver-gences occur between the two, creating periods of risk. The 1983/84correction, the 1987 crash, the 1990 correction, the 1994 correctionand the sell-off in July 1996 all stand out as such periods. As wasdescribed earlier in this paper, the goal here is to identify these peri-ods through the application of technical trend and momentum stud-ies. The product of these studies will be an indicator that gives theappropriate warnings signs when these divergences reach dangerouslevels.

First, however, we should quantify the reliability of E/I by per-forming a linear regression (using the least squares approach) in orderto determine what the correlation is between E/I and the S&P 500.Table 1 shows the output of this regression (using the computer pro-gram "Econometric Views"), and below that is a graph depicting theindependent variable (E/I), the dependent (fitted) variable (S&P500), and the residual.


Table 1

LS//Dependent Variable is S&PDate: 09/14/96 - Time: 14.48

Sample: 5/07/1982 9/13/1996Included observations: 750

Variable Coefficient Std. Error t-Statistic Prob.C -0.037291 0.020610 -1.809403 0.0708E/I 1.035041 0.008461 122.3276 0.0000R-Squared 0.95239 Mean dependent var 2.476477Adjusted R-squared 0.952330 S.D. dependent var 0.197685S.E. of regression 0.043162 Akaike info criterion -6.282943Sum squared resid 1.393470 Schwarz criterion -6.270623Loglikelihood 1293.900 F-statistic 14964.05Durbin-Watson stat. 0.048247 Prob. (F-statistic) 0.000000

The key statistic to look at in Table 1 is “R-squared,” which mea-sures the success of the regression in predicting the values of thedependent variable within the sample. An R2 of 1.0 means that theregression fits perfectly while a reading of 0 means that it fits no bet-ter than the simple mean of the dependent variable. In our regres-sion the R2 is 0.95239, which is an excellent fit. This means that 95pct of the behavior in the S&P 500 can be explained by the behaviorof E/I. This is encouraging, because when we build our trend/mo-mentum indicators, we will have confidence that we are “barking upthe right tree,” as it were.

Chart 5

PART II: THE TECHNICAL INDICATOR

Now that the input has been created and tested for reliability us-ing quantitative analysis, we can get to the juicy part as techniciansand build a risk indicator for the S&P 500. Three traditional techni-cal studies are calculated. Two are momentum studies: a 26-wk rate-of-change (ROC) and a 52-wk slowed stochastic (STOCH). One is atrend study: a 13-wk/26-wk Moving Average Convergence/Divergence(MACD). The time frame for these studies is the intermediate term(3-mo-12-mo), given that the intended audience for this indicator isportfolio managers.

Table 2 shows the one year’s worth of data and formulas. ColumnC shows the weekly closing levels for the S&P 500. Column D showsthe I/B/E/S earnings estimates. Because the series is monthly andour study is weekly, the same number is repeated for all the weeks inany month. Column E shows the 30- year Treasury yield and ColumnF shows our indicator E/I. Columns G through P show the outputfor the above stated studies.The technical studies are calculated as follows:➱ 6 mo ROC: This is a simple rate of change indicator using the

formula:ROC = (E/It26 ÷ E/It1) -1 * 100

For example, the ROC (column G) of E/I (column F) as of 9/13/96(row 52) is [566.20 ÷ 575.26] -1 * 100 = -1.575, meaning that E/I is1.575 pct lower than it was 6 months ago (row 26). This is a usefulmeasure for indicating positive or negative momentum in E/I.

➱ 52-wk STOCH: This is a more complicated momentum measure,and comprises the following calculations: First, we calculate Fast%K (column M) through the following formula:

Fast %K = {E/It52

- min(E/It1:E/I

t52)} ÷ {max(E/I

t1:E/I

t52) -

min(E/It1:E/I

t52)} * 100

Then we calculate the slow %D (column N) by taking a 3-weeksmoothed moving average (SMA) of the fast %D (the MA is smoothedby taking the sum of the previous three fields, subtracting the mostrecent MA, adding the latest value, and dividing the result by 3). Aslow %D (column O) is calculated by taking a 9-week SMA of the Fast%D. Finally, Column P shows the difference between column N andcolumn O in order to indicate whether STOCH is on a buy or sellsignal. A buy signal is given when the fast %D is above the slow %D,and vice versa.

For example, the latest week's fast %K value is (566.2 - 554.6) ÷(632.9 - 554.6) = 14.76. The 3 week SMA is [(52.15+20.67+10.99) -32.51 + 14.76] ÷ 3 = 22.02. The slow %D is 42.93, thus creating a sellsignal. Besides giving buy or sell signals, this study is also useful ingauging whether the absolute momentum level is high or low. Forinstance, a level of less than 50 combined with a sell signal would bequite bearish.➱ MACD: This is a useful trend study which consists of the spread

between two exponentially smoothed moving averages (ESMA)of E/I. The conventional approach is to take 13 weeks and 26weeks as the two M/A's. The 13-week exponentially smoothedM/A (column H) is calculated as follows:

ESMAt14 = ESMAt13 - 0.153846 * (ESMAt13 - E/It14)where ESMAt13 is the first MA in the series and consists of a simple 13week MA. The number 0.153846 is the product of a smoothing con-

Chart 6


A B C D E F G H I J K L M N O PS&P Exp. 30yr 6mo 13wk 26wk 9wk up/ Fast Fast Slow %SD<500 Earn. Tsy E/I ROC ESMA ESMA MACD ESMA down %K %D %D %FD

1 9/21/95 587.3 36.7 6.53 562.3 24.5 538.62 515.49 23.12 22.51 0.61 99.07 99.97 93.62 6.352 9/28/95 588.2 37.2 6.57 566.1 24.8 542.85 519.39 23.46 22.72 0.74 100.00 99.70 94.23 5.473 10/5/95 586.8 37.2 6.45 576.6 24.9 548.05 523.79 24.25 23.06 1.19 100.00 99.79 95.17 4.624 10/12/95 588.2 37.2 6.40 581.1 25.7 553.14 528.20 24.93 23.48 1.46 100.00 99.76 96.16 3.605 10/19/95 589.7 37.2 6.32 588.5 27.3 558.58 532.84 25.74 23.98 1.76 100.00 100.08 97.44 2.646 10/26/95 578.7 37.2 6.35 585.7 26.0 562.75 536.91 25.84 24.39 1.45 98.67 99.53 98.35 1.187 11/2/95 582.5 37.5 6.30 594.6 24.4 567.66 541.35 26.31 24.82 1.49 100.00 99.71 98.99 0.738 11/9/95 594.5 37.5 6.29 595.6 20.5 571.96 545.52 26.43 25.18 1.26 100.00 99.65 99.49 0.179 11/16/95 601.2 37.5 6.27 597.5 19.5 575.88 549.52 26.36 25.44 0.92 100.00 99.67 99.73 -0.0510 11/23/95 601.6 37.5 6.26 598.4 18.1 579.36 553.28 26.07 25.58 0.49 100.00 100.11 99.81 0.3011 11/30/95 608.3 37.9 6.19 612.4 15.7 584.45 557.83 26.61 25.81 0.80 100.00 99.96 99.80 0.1712 12/7/95 618.3 37.9 6.04 627.6 17.7 591.09 563.20 27.89 26.27 1.62 100.00 100.01 99.83 0.1813 12/14/95 622.8 37.9 6.07 624.5 17.8 596.24 567.92 28.32 26.73 1.59 98.64 99.54 99.8 -0.2614 12/21/95 618.4 37.9 6.12 619.4 15.5 599.81 571.89 27.92 26.99 0.93 96.39 98.50 99.66 -1.1715 12/28/95 615.9 38.0 6.00 632.9 18.7 604.91 576.58 28.32 27.29 1.04 100.00 98.85 99.54 -0.7016 1/4/96 616.7 38.0 6.00 632.9 18.5 609.22 580.92 28.30 27.51 0.79 100.00 98.73 99.47 -0.7417 1/11/96 601.8 38.0 6.12 620.5 16.0 610.96 583.97 27.00 27.40 -0.40 94.44 97.37 99.21 -1.8518 1/18/96 611.5 38.0 6.01 631.9 22.8 614.18 587.65 26.53 27.20 -0.68 99.53 98.87 99.15 -0.2919 1/25/96 621.6 38.2 6.06 631.1 23.8 616.79 591.00 25.79 26.89 -1.10 99.19 98.10 98.99 -0.8920 2/1/96 635.8 38.2 6.08 629.1 19.9 618.68 593.92 24.75 26.42 -1.66 98.16 97.74 98.74 -1.0021 2/8/96 656.4 38.2 6.13 623.9 19.4 619.48 596.23 23.25 25.71 -2.46 95.65 98.26 98.58 -0.3222 2/15/96 650.4 38.2 6.11 626.0 19.8 620.48 598.52 21.96 24.88 -2.92 96.58 97.11 98.27 -1.1723 2/22/96 659.1 38.2 6.39 598.5 13.4 617.11 598.52 18.58 23.48 -4.90 82.96 92.08 97.48 -5.4024 2/29/96 644.4 38.4 6.45 595.8 8.4 613.82 598.31 15.51 21.71 -6.20 80.32 87.81 96.38 -8.5725 3/7/96 633.6 38.4 6.47 593.9 6.6 610.76 597.97 12.79 19.73 -6.94 79.35 83.80 94.83 -11.0326 3/14/96 641.4 38.4 6.68 575.3 2.0 605.30 596.23 9.08 17.36 -8.28 68.15 75.66 92.44 -16.7827 3/21/96 650.6 38.4 6.67 576.1 2.5 600.81 594.68 6.13 14.87 -8.73 68.63 73.60 90.07 -16.4728 3/28/96 645.5 38.5 6.65 579.6 2.4 597.56 593.52 4.03 12.46 -8.43 70.28 70.94 87.23 -16.2929 4/4/96 655.9 38.5 6.68 577.0 0.1 594.40 592.26 2.15 10.17 -8.02 67.35 67.82 84.18 -16.3530 4/11/96 636.7 38.5 6.88 560.3 -3.6 589.15 589.80 -0.65 7.76 -8.41 57.40 65.28 80.91 -15.6331 4/18/96 645.1 38.5 6.80 566.9 -3.7 585.72 588.03 -2.31 5.53 -7.84 61.26 63.67 77.43 -13.7632 4/25/96 653.5 38.5 6.79 567.7 -3.1 582.95 586.47 -3.52 3.52 -7.03 61.17 61.17 73.82 -12.6533 5/2/96 643.5 38.9 6.96 559.6 -5.9 579.35 584.40 -5.05 1.61 -6.66 52.67 57.11 70.34 -13.2334 5/9/96 654.9 38.9 7.02 554.8 -6.9 575.57 582.12 -6.55 -0.20 -6.35 43.61 53.87 66.95 -13.0935 5/16/96 671.5 38.9 6.87 566.9 -5.1 574.24 580.95 -6.71 -1.65 -5.06 50.29 51.29 63.72 -12.4236 5/23/96 680.6 38.9 6.84 569.4 -4.9 573.49 580.06 -6.57 -2.74 -3.83 49.66 48.31 61.04 -12.7237 5/30/96 667.0 39.4 6.93 568.3 -7.2 572.69 579.15 -6.47 -3.57 -2.90 47.41 47.55 58.44 -10.8938 6/6/96 673.8 39.4 7.03 560.2 -10.8 570.76 577.69 -6.93 -4.32 -2.62 40.84 46.88 56.06 -9.1739 6/13/96 665.9 39.4 7.09 555.4 -11.1 568.40 575.98 -7.58 -5.04 -2.54 36.99 42.67 53.53 -10.8640 6/20/96 666.8 39.4 7.10 554.6 -10.5 566.29 574.34 -8.05 -5.71 -2.34 36.35 39.64 50.96 -11.3241 6/27/96 670.6 39.5 6.89 573.1 -9.5 567.33 574.24 -6.91 -5.98 -0.94 51.31 41.95 48.83 -6.8842 7/4/96 675.9 39.5 7.00 564.0 -10.9 566.82 573.46 -6.63 -6.12 -0.51 43.99 42.23 46.96 -4.7343 7/11/96 651.0 39.5 7.11 555.3 -10.5 565.05 572.06 -7.01 -6.32 -0.69 36.90 42.11 45.50 -3.4044 7/18/96 638.7 39.5 7.00 564.0 -10.7 564.90 571.44 -6.55 -6.37 -0.18 43.99 44.70 44.65 0.0545 7/25/96 635.9 39.5 7.02 562.4 -10.9 564.52 570.75 -6.23 -6.34 0.11 42.69 40.96 43.59 -2.6446 8/1/96 662.5 39.9 6.94 574.3 -8.7 566.02 571.02 -5.00 -6.04 1.04 46.86 43.16 43.14 0.0247 8/8/96 664.1 39.9 6.75 590.4 -5.4 569.77 572.51 -2.74 -5.31 2.57 61.50 50.63 43.53 7.0948 8/15/96 663.1 39.9 6.77 588.7 -6.0 572.68 573.76 -1.08 -4.37 3.29 59.92 53.45 44.22 9.2349 8/22/96 667.0 39.9 6.84 582.7 -2.7 574.21 574.44 -0.23 -3.45 3.22 52.15 55.66 45.58 10.0850 8/29/96 660.9 39.9 7.03 566.9 -4.8 573.09 573.86 -0.77 -2.85 2.08 20.67 46.19 46.16 0.0351 9/5/96 653.9 40.0 7.11 563.3 -5.2 571.58 573.05 -1.47 -2.55 1.08 10.99 32.51 45.05 -12.5352 9/12/96 671.1 40.0 7.07 566.2 -1.6 570.75 572.52 -1.77 -2.37 0.60 14.76 22.02 42.93 -20.91

Table 2

stant 2 divided by the M/A period of 13. The 26 week ESMA is calcu-lated in similar fashion (column I). The MACD is a simple spreadbetween the 13-wk ESMA and 26-wk ESMA (column J). To createbuy and sell signals, a 9-wk ESMA is taken of the spread (column K).Finally, column L shows the difference between columns J and K,indicating whether MACD is on a buy or sell signal. Also, the abso-lute level of MACD is important to identify the magnitude of risingand falling trends.

For example, the latest value for the 13-wk ESMA is 571.58 -(0.153846 * (571.58 - 566.2) = 570.75. The 26-wk ESMA is 571.58 -

(0.076923 * (573.05 - 566.2) = 572.52. The MACD is the spread:570.75 - 572.52 = -1.77. The ESMA is -2.37, creating a sell signal.

Chart 6 depicts these studies. Chart 6 nicely shows what happensto these indicators when E/I gets into the danger zone as a valuationmodel for the S&P 500. The major periods of correction/consolida-tion in the stock market all were signaled by negative readings in theROC, MACD and STOCH. However, eyeballing these studies to de-termine risk in the S&P 500 is not very scientific, and a more system-atic approach is needed. We accomplish this by establishing certainconditions on the three technical studies. The most straightforward


approach is to define an "if-then" condition for each study, and thencombine the results into a composite trend/momentum signal.

Table 3

The conditions will be very simple so we can determine if there ismethod that works (i.e. gives reliable signals) without having to getinto back testing and optimization.■ ROC SIGNAL: For ROC, we simply tell the computer to return

“TRUE” if the latest reading is below zero, that is E/I is below itslevel of 6 months ago. Otherwise, return “FALSE.” The results

are depicted in Table 3 in column E.■ STOCH SIGNAL: Here we need to add a twist to account for the

fact that this indicator can not only be on a buy or sell signal, butcan also be overbought or oversold. Therefore, we tell the com-puter to return “TRUE” if STOCH is on a sell (i.e. the slow %D isbelow the fast %D) AND STOCH is below 50, indicating that mo-mentum is below neutral (STOCH oscillates between zero and100). If neither of these conditions is met, the computer returns“FALSE.” Column F shows the results.

■ MACD SIGNAL: For MACD, we also set two conditions: return“TRUE” if MACD is below zero, AND it is on a sell signal, meaningthat MACD is below its 9 week ESMA. Otherwise, return “FALSE.”Column G shows the output.Finally, we set a last if-then condition to return “TRUE” when all

three individual conditions are true. If only some are true, or if noneare true, return “FALSE.”

A “TRUE” therefore will signal those time periods when all threetechnical studies tell us that the underlying trend and momentumconditions of our indicator E/I are reaching dangerous levels for theS&P 500.

Given that the stock market can ignore rising rates and deterio-rating earnings momentum for some time without correcting (as wasthe case in 1987), it is important to note again that the objective ofthis signal is to flash a warning to get out of stocks (by hedging) whenE/I's trend and momentum conditions get really dangerous, ratherthan every time the slightest negative divergence occurs.

Bringing all of this together, Chart 7 shows the S&P 500, our indi-cator E/I and those periods of risk as defined by our trend/momen-tum signal. As has been the case with all charts, the S&P 500 and E/I are charted as a log in order to show price changes in equal propor-tion.

A B C D E F G HS&P 500 E/I ROC STOCH MACD COMP

1 9/21/95 587.3 562.3 FALSE FALSE FALSE FALSE2 9/28/95 588.2 566.1 FALSE FALSE FALSE FALSE3 10/5/95 586.8 576.6 FALSE FALSE FALSE FALSE4 10/12/95 588.2 581.1 FALSE FALSE FALSE FALSE5 10/19/95 589.7 588.5 FALSE FALSE FALSE FALSE6 10/26/95 578.7 585.7 FALSE FALSE FALSE FALSE7 11/2/95 582.5 594.6 FALSE FALSE FALSE FALSE8 11/9/95 594.5 595.6 FALSE FALSE FALSE FALSE9 11/16/95 601.2 597.5 FALSE FALSE FALSE FALSE10 11/23/95 601.6 598.4 FALSE FALSE FALSE FALSE11 11/30/95 608.3 612.4 FALSE FALSE FALSE FALSE12 12/7/95 618.3 627.6 FALSE FALSE FALSE FALSE13 12/14/95 622.8 624.5 FALSE FALSE FALSE FALSE14 12/21/95 618.4 619.4 FALSE FALSE FALSE FALSE15 12/28/95 615.9 632.9 FALSE FALSE FALSE FALSE16 1/4/96 616.7 632.9 FALSE FALSE FALSE FALSE17 1/11/96 601.8 620.5 FALSE FALSE FALSE FALSE18 1/18/96 611.5 631.9 FALSE FALSE FALSE FALSE19 1/25/96 621.6 631.1 FALSE FALSE FALSE FALSE20 2/1/96 635.8 629.1 FALSE FALSE FALSE FALSE21 2/8/96 656.4 623.9 FALSE FALSE FALSE FALSE22 2/15/96 650.4 626.0 FALSE FALSE FALSE FALSE23 2/22/96 659.1 598.5 FALSE FALSE FALSE FALSE24 2/29/96 644.4 595.8 FALSE FALSE FALSE FALSE25 3/7/96 633.6 593.9 FALSE FALSE FALSE FALSE26 3/14/96 641.4 575.3 FALSE FALSE FALSE FALSE27 3/21/96 650.6 576.1 FALSE FALSE FALSE FALSE28 3/28/96 645.5 579.6 FALSE FALSE FALSE FALSE29 4/4/96 655.9 577.0 FALSE FALSE FALSE FALSE30 4/11/96 636.7 560.3 TRUE TRUE FALSE FALSE31 4/18/96 645.1 566.9 TRUE TRUE FALSE FALSE32 4/25/96 653.5 567.7 TRUE TRUE FALSE FALSE33 5/2/96 643.5 559.6 TRUE TRUE FALSE FALSE34 5/9/96 654.9 554.8 TRUE TRUE FALSE FALSE35 5/16/96 671.5 566.9 TRUE TRUE FALSE FALSE36 5/23/96 680.6 569.4 TRUE TRUE FALSE FALSE37 5/30/96 667.0 568.3 TRUE TRUE FALSE FALSE38 6/6/96 673.8 560.2 TRUE TRUE FALSE FALSE39 6/13/96 665.9 555.4 TRUE TRUE FALSE FALSE40 6/20/96 666.8 554.6 TRUE TRUE FALSE FALSE41 6/27/96 670.6 573.1 TRUE TRUE TRUE TRUE42 7/4/96 675.9 564.0 TRUE TRUE TRUE TRUE43 7/11/96 651.0 555.3 TRUE TRUE TRUE TRUE44 7/18/96 638.7 564.0 TRUE TRUE FALSE FALSE45 7/25/96 635.9 562.4 TRUE FALSE TRUE FALSE46 8/1/96 662.5 574.3 TRUE FALSE FALSE FALSE47 8/8/96 664.1 590.4 TRUE FALSE FALSE FALSE48 8/15/96 663.1 588.7 TRUE FALSE FALSE FALSE49 8/22/96 667.0 582.7 TRUE FALSE FALSE FALSE50 8/29/96 660.9 566.9 TRUE FALSE FALSE FALSE51 9/5/96 653.9 563.3 TRUE FALSE TRUE FALSE52 9/12/96 671.1 566.2 TRUE FALSE TRUE FALSE

Chart 7


Date Portfolio Value Value of # Value of Initial Transaction Margin Finance Cumulative Weekly Portfolio Value-unhedged 1 contract contracts hedge Margin Commission Adjustment Charges Drawdown Pct P/L -Hedged

6-Apr-84 11,333,853 SELL 11,333,85313-Apr-84 11,309,883 78,135 145 11,329,575 -1,812,500 -2,320 -0.02% -0.02% 11,307,56320-Apr-84 11,466,629 79,150 11,476,750 -1,812,500 -147,175 -690 -1.33% -1.31% 11,316,41227-Apr-84 11,508,444 79,370 11,508,650 -1,812,500 -31,900 -690 -1.61% -0.29% 11,325,0884-May-84 11,666,897 80,395 11,657,275 -1,812,500 -148,625 -690 -2.93% -1.32% 11,331,701

11-May-84 11,599,331 79,860 11,579,700 -1,812,500 77,575 -690 -2.25% 0.68% 11,342,96118-May-84 11,424,371 78,585 11,394,825 -1,812,500 184,875 -690 -0.63% 1.62% 11,356,05325-May-84 11,125,815 76,460 11,086,700 -1,812,500 308,125 -690 2.08% 2.71% 11,366,717

1-Jun-84 11,004,428 75,555 10,955,475 -1,812,500 131,225 -690 3.23% 1.15% 11,373,2368-Jun-84 11,269,946 77,310 11,209,950 -1,812,500 -254,475 -690 0.98% -2.24% 11,392,487

15-Jun-84 11,042,683 75,680 CLOSE 10,973,600 -1,812,500 236,350 -690 3.05% 2.07% 11,398,413

Table 4

PART III: IMPLEMENTING THE HEDGING STRATEGY

Before evaluating the success of our hedging strategy, certain as-sumptions need to be set regarding its implementation.1. The study beginning with a portfolio of $10,000,000 in January

1983. That date reflects the beginning of the various technicalstudies that are in use. Since they are trend and momentum based,there is by definition a lag between the beginning of the indicatorE/I and its studies STOCH, ROC, and MACD.

2. The cash portfolio is indexed to the S&P 500 index, and the tim-ing signal is executed through the short sale of S&P index futures.

3. The timing model is updated every Friday afternoon at 3:30 pmto allow for enough time for any transactions to be executed inthe market as of that week. This is not entirely scientific becausethe indicator historically reflects end-of-week values, but it will haveto do for the purpose of this study. Waiting for the next Mondayleaves too large a gap in terms of potential price swings.

4. When a signal is generated (i.e. when all three study conditionsare TRUE), a transaction is immediately executed consisting ofthe short sale of S&P 500 futures equal to the total market valueof the portfolio at that time. This is done as follows: The marketvalue of the portfolio is divided by the product of $500 and theprice of the S&P 500 index. An example illustrating the first sig-nal in 1984 is shown in the table below. On Friday, April 13th,1984, a signal is given. At that time, the value of one S&P contractis $500 x 156.27 = $78,135. The number of contracts needed tohedge the portfolio is therefore 145 (11,309,853 ÷ 78,135).

5. Transaction costs associated with the short sale are as follows: Thecommission is $16 for a round-trip, and a financing spread of 2percentage points is applied to the cost of the initial margin. Thequestion of margin is tricky, because if cash on hand is available,then by definition the entire portfolio is not invested in the S&P500. Therefore, for simplification's sake, it will be assumed thatthe margin is borrowed at a cost of 2 pct over and above what themargin amount will earn in the futures account. Additional mar-gin will be applied directly to the cash account, however. Table 4shows the cumulative drawdown/profit for the hedge.

6. When the signal ends (when the condition is FALSE again), the145 contract short sale is covered. However, this will not be knownuntil the end of that week when the closing figures are inputtedinto the model. Therefore, when looking at the history of signals,the P&L calculations start with a one week lag and continue foran extra week. The final column shows the portfolio value on ahedged basis.Table 5 gives the details of all the signals: each observation with

start and end date, the number of weeks that the signal is in effect,

the start and end values for the S&P 500, the percentage change inthe S&P 500, and finally the total effect of hedging the S&P 500 port-folio with futures contracts.

PERFORMANCE

There have been 9 observations, each of which has lasted any-where from 3 weeks to 11 weeks. The average period lasted about 7weeks. Six out of nine signals were profitable, or 67 pct. The biggestgain from hedging occurred during the 1987 crash (26.13 pct), whilethe largest drawdown occurred in 1990, totaling 5.75 pct. The aver-age gain from hedging is 4.57 percentage points. The difference be-tween the change in the portfolio value and the return of the hedgecan be attributed to transaction costs and the rounding up or downof the number of contracts that need to be sold short.

Table 5

9 Observations Summary of ObservationsChg in Pct P/L of

Date # Weeks S&P 500 S&P 500 Hedge#1 begin 4/13/84 156.27

end 6/15/84 9 151.36 -3.14% 3.05%#2 begin 4/10/87 322.36

end 6/5/87 8 292.06 -9.40% 9.29%#3 begin 9/4/87 322.36

end 10/30/87 8 238.14 -26.13% 26.43%#4 begin 3/16/90 338.30

end 5/25/90 10 357.74 5.75% -5.75%#5 begin 8/24/90 317.10

end 10/19/90 8 303.83 -4.18% 4.13%#6 begin 4/24/92 410.17

end 5/15/92 3 414.89 1.15% -1.16%#7 begin 5/6/94 451.39

end 7/22/94 11 453.28 0.42% -0.45%#8 begin 10/7/94 455.46

end 11/25/94 7 452.65 -0.62% 0.59%#9 begin 6/28/96 668.18

end 7/19/96 3 634.91 -4.98% 4.96%

Average 7.4 -4.57% 4.57%ex-1987 7.4 -1.88% 1.83%

Since the beginning of this study in 1983, the average annual totalreturn of a buy-and-hold portfolio has been 15.90 pct, while the aver-age annual return of an actively hedged portfolio using this timingindicator has been 19.64 pct. Hence, the average yearly excess returnis 3.74 percentage points.


Chart 8

Chart 8 shows the cumulative total return of our hypothetical port-folio on an actively hedged basis. The dotted line shows the totalreturn on a buy-and-hold basis.

ANALYSIS

A few issues arise when we look at the results.■ The biggest problem is immediately apparent when looking at

the bottom two rows of the table as well as the chart with the cu-mulative total returns: while it is desirable that our indicator cap-tured the 1987 stock market crash, the problem is that it accountsfor a big part of the overall profitability. Including the crash, theaverage excess return is 4.57 pct, but excluding the crash, the ex-cess return is only 1.83 pct. However, from the standpoint of elimi-nating market risk from a portfolio from time to time, this is stillan acceptable performance (because all we’re giving up is upsideperformance, as opposed to being outright short).

■ The study only includes the bull market of the 1980s and 90s, andtherefore we are unsure whether it will stand the test of bull andbear markets. The problem is that we want a forward-lookingindicator and are limited by the availability of earnings estimates,leaving only 1982 and after. However, we can go back farther intime to assess the validity of the unorthodox discounting methodusing actual earnings instead of expected earnings. If the corre-lation of the indicator stands up over a longer period of time(through the same regression analysis as before), we can at leastascertain that the fundamental idea behind E/I is valid. Chart 9shows E/I using expected earnings after 1982 and actual earningsbefore 1982. The regression study (not shown) reveals an R2 of92 pct, which is still pretty good for a period of 35 years. As aresult, we can maintain confidence that the idea behind the E/Iindicator is valid.

■ Chart 7 on shows that while the signal captured some correctionsperfectly (namely the 1987 crash and the sell-off in July ’96), ithas been on the late side in other instances, such as 1990 and1994. It appears that the sharp price corrections are handled wellby our indicator, while the more triangle shaped time-based cor-rections are handled with less success. By the time the signalsoccur in the latter type corrections, it seems a better time to buythan to sell. The problem is that if the three studies used for thisexercise are made more sensitive (by giving an earlier signal), thesharp advances prior to the 1987 crash and July ’96 sell-off arehedged out, leaving profits on the table. The latter point is a validone, and the performance table does shows a rather large maxi-mum drawdown of 5.75 pct in 1990. One way to improve theeffectiveness of the signal is to try different indicators and to runthem through a computer spreadsheet (Microsoft Excel was usedfor this study).

■ The indicator has not only been adept in signaling risk periods,but was very effective in signaling market bottoms as well. Thechart shows that while E/I peaks well in advance of the S&P 500index, it bottoms at the same time as the index. In other words, itis a leading indicator at tops and a coincident indicator at bottoms.

■ Relating to the previous point, we see that E/I was also valuable asa confirming indicator of a rising stock market. Knowing that E/I tended to peak well in advance of the stock market, we can as-sume that as long as E/I is in its uptrend (making new highs, up-trend lines intact, etc.), it is safe to be aggressively invested in stocks(note the 1995 period). In other words, the indicator workedboth ways. When both stocks and E/I are rising and making newhighs, stay invested. When the bearish divergence first occurs, itis OK to remain invested, but caution is warranted. As the diver-gence gets progressively worse and the retracement of E/Is pre-ceding advance deepens, it is time to get ready to hedge. Whenthe timing model kicks in, or when other technical studies indi-cate risk, the S&P is sold, at which point the end of the signal canbegin to be anticipated (in terms of the percentage retracementand the time of the correction). Finally, the signal ends, and theS&P is bought back.

CONCLUSION

We know that the correlation of E/I is 95 pct, meaning that 95 pctof the action in the stock market has been explained by E/I. That ispretty valuable. Therefore, if we compliment the timing indicatorwith traditional technical analysis (on both E/I as well as the stockmarket itself), perhaps we can increase its value in identifying riskyperiods in the stock market, as well as periods during which a fully-invested portfolio stance should be adopted. Complementing ourindicator with traditional measures such as the Advance-Decline Line,Lowry’s Buying Power, Cash Flows into mutual funds and other senti-ment measures should nicely round its effectiveness.

One example of performing additional ad hoc analysis on E/I isshown in Chart 10. There have been four major corrections in E/I(excluding the 1996 decline), each of which retraced between 46 pctand 52 pct of the preceding advance. The average correction is 48pct. This falls right in the middle of the traditional Fibonacciretracement objectives of 38.2 pct, 50.0 pct, and 61.8 pct. Note alsothat out of those four corrections, three were doubles (two sell sig-nals). Hence, there is a repeated pattern evident in the behavior ofE/I, and that can be very valuable. For instance, we can deduce thatthe July ’96 correction was perhaps only the first of two correctionphases, and that the stock market will remain at risk until E/I cor-rects by the 48 pct average. It will be up to the technical strategist to

Chart 9


combine his or her skills with the knowledge that E/I is a valid lead-ing indicator of stock market peaks. Anything from trendline analy-sis to time cycles may be of value here.

The indicator developed in this paper, E/I, can be of use wheninvesting in the stock market in several ways. For portfolio managersand position traders alike, utilizing a fundamentally oriented indica-tor with a 95 pct correlation to stock prices is useful, either throughthe use of a timing indicator as done in this paper, or merely as anindicator of risk in combination with other indicators. The indicatorworks both as a risk indicator and as a fully invested indicator. Whilethis paper has focused on portfolio managers who can use E/I tohedge during high risk periods, obviously it can be equally valuableto position traders who can use stop-reverse strategies for either along-neutral strategy or a long-short strategy. Written in August 1996.

Chart 10

UPDATE - ONE YEAR LATER

Updating the model from the report’s initial publication in Au-gust 1996, we find that no further sell signals have been issued sincethe June/July 1996 signal that captured that sell-off so well. Whilethe indicator E/I did not issue a sell signal going into the April 1997correction, a more subtle warning was evident in that a bearish diver-gence occurred at the final high leading into the correction. Thisreinforces my point that the value of E/I as an indicator is not lim-ited to a buy/sell algorithm, but rather that traditional technical analy-sis can be performed just as would be appropriate for an advance-decline line for example. Since the correction in April ’97 was soshort lived, in retrospect the failure of E/I to issue a sell signal prob-ably worked out for the better. Since the April 14th low, the indica-tor has been making new highs, thereby confirming the bullish priceaction. August 18, 1997.

Chart 11

REFERENCES

■ Paper written by Paul Montgomery for 1996 MTA seminar wherehe was the featured speaker

■ John Murphy, Technical Analysis of the Futures Markets■ CQG, Inc.■ Welles Wilder, New Concepts in Technical Trading Systems

BIOGRAPHY

Jurrien Timmer, CMT, was born and raised in Aruba, andcame to the U.S. to attend Babson College in 1981. He gradu-ated with a B.S. in finance & investments and went to work forABN/AMRO Bank, where he worked for 10 years and ended uprunning a fixed income trading & sales desk. It was there hedeveloped an interest in technical analyis, and started to write aweekly report for institutional clients. His interests changed fromtrading/sales to research. Currently he is a senior technical ana-lyst at Fidelity Investments, and advises the investment profes-sionals on developments in the financial markets (stocks andbonds, as well as sectors and industry groups). He applies quan-titative methods to back up his technical ideas, since he believesdrawing trend lines is not enough to add any value to the invest-ment process. He writes several in-house reports and holdsmonthly technical reviews for the investment professionals.


THE STORY OF THE THREE STOCK MARKET BOTTOMS:The Papa Bottom, The Mama Bottom and The Baby Bottom

Kenneth Safian

9The most unique aspect of this stock market that many investors

may be telling their children or grandchildren is the three distinctstock market bottoms that have occurred since March 2000. As page7 illustrates, more than 50% of the Standard & Poor’s 500 Indexgroups reached new 52-week lows at the 1987 and 1990 bottoms. Justunder 50% reached new lows in 1998. In April 2000, 50% of thegroups reached new lows as our Technology Average had its finaladvance. Investors, therefore, dramatically reduced their more con-servative holdings and heavily purchased technology company shares.This was the larger sector bottom for the market since more groupsreached their lows at that time.

The second bottom occurred in October 2000 when worse eco-nomic statistics were released and investors heavily sold their cycli-cally sensitive issues. At that time, 25% of the Standard & Poor's 500Index groups reached new lows. This sector low was discountingunfavorable economic conditions and the liquidation was for funda-mental rather than psychological reasons as was the case in the ear-lier bottom. Finally, the technology issues reached their lows thisMarch when the media declared Lucent and some other corpora-tions being close to bankruptcy. Fifteen percent of the S&P 500 In-dex groups hit new 52-week lows at that time. This was a larger psy-chological bottom due to short selling, but a smaller actual bottomsince fewer groups reached new lows. These stocks were no longer asimportant to the total market because they had already declined somuch in dollar values. The media focused on this bottom as being“the” bottom because of the interest in technology issues.

The table below shows the performance of some industry groupsduring these time periods. These different bottoms in price can alsobe seen for the breadth series we maintain for our averages (see charts10 to 14).

Safian Investment Research AveragesFirst Bottom

Group Bottom Rally Period % ChangeDate Price Date Price

Food Average 3/17/00 564.2 7/21/00 815.8 44.6%

Life Insurance 3/17/00 364.0 6/2/00 543.6 49.3%

Property & Cas. 3/10/00 455.1 5/26/00 755.6 66.0%

Electric Utility 3/10/00 732.8 4/28/00 908.5 24.0%

Drugs 3/24/00 2458 7/14/00 3633 47.8%

Second BottomGroup Bottom Rally Period % Change

Date Price Date Price

Cyclical Average 10/20/00 327.9 3/9/01 480.4 46.5%

Capital Goods Avg. 10/20/00 258.3 3/2/01 379.6 46.9%

Paper 10/20/00 520.2 1/5/01 822.0 58.0%

Retail 10/31/00 553.7 2/2/01 867.6 56.7%

Third BottomGroup Bottom Rally Period % Change

Date Price Date Price

Technology Avg. 4/6/01 322.8 4/27/01 485.4 45.9%

Net Working 3/16/01 1336.5 4/20/01 1994.5 45.9%

Semiconductor 4/6/01 661.7 5/4/01 1102.9 66.7%

Copyright © 2001 by Safian Investment Research, Inc. www.safian.com

This perspective brings up some interesting points. First, the Fed-eral Reserve eased dramatically at the end of 1999 and the beginningof 2000 fearing a year 2K problem could occur. Additional fundswere available in the system and investors pushed up the prices ofmore aggressive (technology) stocks compounding the “bubble.” Ayear-over-year gain of more than a 150% occurred for our Technol-ogy Average in early 2000 while industrial production for technologyindustries grew at a peak annual rate of gain of 60%. Prior to that,both series grew about the same. It is noteworthy to re-emphasizethat most S&P 500 Index groups reached new lows at the same timethat our Technology Average reached its peak.

The diversity within the economy, the fiscal drag created by thelarge budget surplus of our government during the mid-to-late 1990s,and the raising of interest rates by the Federal Reserve when the busi-ness conditions were “too strong” also strengthened the dollar. Theseconditions were quite unfavorable for commodity cyclical industries.Despite those trends, the price of energy was able to rise due to theenergy policies of our government and those comprising OPEC.Given these trends, why should there not have been diversity withinthe economy and stock market and why should they not continue?Finally, if all groups reached their lows, had their good bounces, andthe first bottom occurred over one year ago, why should there be anurgency to rush out and buy stocks generally? The technology sectormay rally further since that group reached its low most recently.However, most groups backed off after the rally from their bottomwas over and we would expect the same situation for technology stocksonce their rally is exhausted. Additionally, the rebounds in groupsfollowing their lows were about 45% and our Technology Averagehas rallied that percentage.

The graphic section of our recent study highlights the currenttechnical and fundamental environment. More technical indicatorssuggest an approaching overbought condition. For example, whilemore groups were oversold than overbought several weeks ago, thereis now an equal number if both the intermediate and long term cat-egories are included. Most recent oversold conditions have beenoffset by overbought conditions indicating a neutral environment.Investor sentiment data have generally moved to more optimistic at-titudes but they are not yet at dangerous levels.

Mutual Fund data clearly reflect the divergent trends discussedearlier in this report. There were relatively large net conversions outof aggressive growth funds in February and March, but growth andincome funds were getting net inflows. Money managers were alsonet buyers of stocks and net sold securities other than common stocks


which did not indicate bearish attitudes by the mutual fund manag-ers. Net flows into growth and income funds continued in March2001. From a fundamental point of view, March inventories of tech-nology products still seemed high and new orders of technology prod-ucts fell below year ago levels for the first time since 1991. The datafor the Purchasing Managers’ Survey seemed mixed and may suggestsome stabilization. Employment data for April seemed more diversethan generally expressed in the media. For example, the percentageof industries that recorded a greater number of workers on a month-to-month basis has stabilized over the past three months. Structuralchanges seem to be taking place in our labor force since the numberof women workers are falling and men are rising slightly. Addition-ally, there have been large declines in employment in the personnelagencies category.

We continue to believe the government is and will be the majordeterminant of economic and sector trends. Congress is now in theprocess of trying to pass a budget resolution and the Republican lead-ership must compromise in order not to have that resolution defeatedin the Senate. A defeat of a budget resolution in the Senate wouldmean that future budget resolutions could be filibustered. It seemsthe government monetary and fiscal stimulus may abort the reces-sion signal given by our Composite Forecasting Index and relateddata in August 2000 and that greater inflationary pressures shouldunfold. Our investment policy is unchanged. We would be morediversified but concentrate more in those companies that will ben-efit from increased government fiscal stimulus and somewhat higherinflation rates. Portfolios should be structured in a similar manneras our Suggested Equity Portfolio for Large Institutional Accounts.

SECTOR ANALYSIS

Sector analysis is an approach that Ken Safian and Ken Smilen(who died several years ago) formally developed in the early 1960s.They developed the Dual Market Principle which first divided thestock market into two major groups – traditional growth and cyclical– and several satellite sectors which included primarily regulated com-panies or ones strongly influenced by the government such as air-lines, utilities, defense, oil and gas. They maintained many funda-mental and technical series for these two major groups just as techni-cians and analysts kept them for the entire market or economy: price,volume, breadth, short interest, odd lot series, dividend yields, P/E’s, valuation measures, etc. Some of these indicators can no longerbe maintained because data are either unavailable or no longer rel-evant. However, other series and relationships are available and aremaintained.

In the mid-1960s when increased government stimulus and com-ing financial speculation, seemed probable an aggressive growth av-erage started, due to a slowing economy. This index proved to achieveextraordinary price gains almost equaling those of the late 1990s (seetable). That average was discontinued because there were so manymergers and restatements of earnings.

Components of the Smilen & SafianAggressive Growth Average in 1968

Yearly Percentage Changes of the Averages

TraditionalAggressive Growth Cyclical S&P

Components Date Growth Average Average 500 Dow

American Hospital Supply Co.

AMP, Incorporated

Automatic Retailers of America, Inc. 1957 +14.7 -20.2 -14.3 -12.8

Avon Products, Incorporated 1958 +65.5 +37.6 +38.1 +33.9

Control Data Corporation 1959 +44.0 +16.8 + 8.5 +16.4

EG&G, Incorporated 1960 +25.8 -18.5 - 3.0 - 9.3

Hewlett-Packard Company 1961 +36.5 +24.3 +18.2 +23.1 +18.7

I.B.M. Corporation 1962 -26.1 -26.3 -12.2 - 11.8 - 10.8

Itek Corporation 1963 +81.7 +21.1 +18.2 +18.9 +11.8

Litton Industries, Incorporated 1964 + 0.1 + 5.7 +12.0 +13.0 +13.9

Perkin-Elmer Corporation 1965 +68.2 +29.1 +11.2 + 9.1 +10.9

Polaroid Corporation 1966 +18.4 + 4.1 - 25.1 - 13.1 -18.9

Sanders Associates, Inc. 1967 +69.7 +33.4 +17.5 +20.1 +15.2

Syntex Corporation 1968 - 4.8 + 0.7 +12.8 + 7.7 + 4.3

Teledyne, Incorporated

Texas Instruments, Inc.

Xerox Corporation

Zenith Radio Corporation

In 1971, when President Nixon instituted wage and price controls,it appeared there would be a distinct difference in earning and priceperformances between consumer related companies and capital goodscompanies. Two new averages to track these sectors were immedi-ately started. The performance between the Consumer Related Av-erage and the Capital Goods Average was tremendous in 1973: Theformer average was down more than 38% and the Capital Goods Av-erage was up 22%. Divergences between sectors are nothing new. In1957, our Cyclical Average declined more than 20% while the Tradi-tional Growth Index increased almost 15%. In our judgement, thecurrent period is a magnification of the sector work started by ourfirm many years ago and the great degree of change has been causedby the transition within our economy. Our firm now has about 50individual sector averages for which prices, volume, breadth and otherrelated series are maintained. Our Technology Average data wereextremely helpful in detecting the deteriorating technical conditionsin early 2000.

Breadth of the stock market can be kept in a number of ways, butthe most common method is to merely compute the difference be-tween the number of advances and declines, on a daily or weeklybasis and cumulate those figures. If the trend in these breadth seriesvary from price or the direction of that trend changes, investors candetect transitions within the stock market. Breadth series for most ofour sector averages have been very helpful in detecting weaker orstronger technical conditions for individual groups of stocks.


CONCLUSION

The important question regarding the recent, unique occurrenceof the differing bottom periods for groups of stocks is whether thiswill develop again. Will it become a more common technical condi-tion at both market tops and troughs? We believe the answer is: “prob-ably so.” As economic conditions become more complex and worldbusiness characteristics differ, we believe there are increased chancesfor major disparities within the stock market to occur. Furthermore,as more funds for the stock market are in retirement accounts, thereare reduced probabilities for these savings to be withdrawn from thestock market, as are regular savings that can go toward the purchaseof a house or a car. Therefore, money stays in these accounts and themoney managers move funds from one group or type of stock to an-other. Given these factors, we think varied sector trends will becomean increasing probability and should be a major part of technicalanalysis.

BIOGRAPHY

Kenneth Safian is President of Safian Investment Research, afirm that has provided investment strategy to major institutionalinvestors here and abroad for almost forty years.

Ken began his investment career at Dreyfus & Co. in 1958where he was responsible for the Investment Management Divi-sion. He co-authored a text in 1960, which first introduced tothe investment community the relationship between growth andcyclical stocks. His firm has been referred to as a “think tank”since it has originated many new investment and economic con-cepts.

Ken is a graduate of the Wharton School at the University ofPennsylvania, a past director and member of the New York Soci-ety of Security Analysts, a member of the National Association ofBusiness Economists, a member of the Investment Policy Com-mittee of Edward D. Jones & Company and formerly served onthe Special Firms Committee of the New York Stock Exchange.He has been a member of the MTA since 1977. Ken has spokenat numerous investment and economic forums, is frequentlyquoted in the media and continually consults with various gov-ernment officials in Washington, D.C.

Please see over for the Breadth Series Charts


ABSTRACT

Purpose: Previously, we have shown that the theoretical returnsfor a simple non-directional option strategy initiated after a suddenand significant volatility implosion of an underlying stock has a posi-tive expectation with an average return per trade of 4.25%. Thisstudy was designed to evaluate whether the addition of a second en-try signal based on the average-true-range (ATR) of the daily stockprices could improve the theoretical returns or decrease the draw-downs experienced in the first study.

Methods and Materials: The 30 Dow Jones Industrial Averagestocks from November 1, 1993 through May 30, 1998, were chosenfor this study. Delta neutral/gamma positive straddle options posi-tions were initiated on the opening price of the stock after two se-quential signals were satisfied. The first signal was generated whenthe near-term historical volatility of the stock had significantly im-ploded relative to its longer-term historical volatility. The secondsignal was generated when the daily ATR of the stock began to in-crease. Any signals generated in the same stock before the 6-weektermination date of a prior trade were ignored. On the date of calcu-lation, the options prices were determined with the actual impliedvolatility using the Black-Scholes model, assuming moderate slippage.All trades were equally weighted. The value of the options positionswere calculated based on the closing stock price at the 2-, 4-, and 6-week periods respectively. Two trading systems were evaluated. Inthe first system (time based system), time was the sole determinantused to determine when the option positions would be closed out.In the second trading system (money management system), simplemoney management rules were added to reduce drawdowns and to“lock-in” profits in profitable trades. Given the wide variability ofbrokerage fees, the results are presented without commission costsdeducted.

Results: A total of 230 trades were generated between 11/1/93and 5/30/98. For the time-based trading system (trading system 1),the 2-week, 4-week, and 6-week cumulative return was +88%, +255%,and -151% and the average return per trade was +0.38%, +1.11%,and -0.65% respectively. For the money management trading system(trading system 2), the 2-week, 4-week, and 6-week cumulative returnwas +219%, +979%, and +1470% and the average return per tradewas +0.94%, +4.26% and +6.39% respectively. The use of a simplemoney management system significantly reduces the drawdowns ofthe system.

Conclusions: The addition of a second entry filter, ATR, did notimprove the theoretical returns of a simple time-based volatility trad-ing strategy that previously had been shown to produce a positivereturn for positions held four weeks. The addition of the ATR filterdid, however, significantly decrease the drawdowns that precludedthe original system’s viability as a useful trading strategy in its ownright. As in the original study, the addition of some simple moneymanagement rules had a dramatic impact on the results. The addi-tion of the ATR filter in combination with some very simple money

management rules significantly improved the theoretical returns whilesimultaneously decreasing the drawdowns when compared to theoriginal study. This improved volatility-based, market-neutral, delta-neutral (gamma positive) trading strategy yielded a very substantialpositive return across a large number of large-cap stocks and across abroad 5-year period. These results demonstrate the potential posi-tive returns that can be obtained from a market-neutral/delta-neu-tral strategy. The benefit of a market-neutral strategy as demonstratedhere is of significant importance to institutional portfolio managersin search of non-correlated asset classes.

INTRODUCTION

For options-based trading, the price action of any freely-tradedasset (e.g., stocks, futures, index futures, etc.) can be grouped intothree generic categories (however defined by the trader): (a) bull-ish price action; (b) bearish price action; (c) congestion/tradingrange price action.

Specific options-based strategies can be implemented which re-sults in profits if any two out of the three outcomes unfold. For ex-ample, the purchase of both call and put options on the same under-lying asset for the same strike price and same expiration date is termeda “straddle” position (e.g., buying XYZ $100 strike March 1999 calland put options = XYZ $100 March 1999 straddle). This straddleposition can be profitable if either (a) or (b) quickly occur withsignificant magnitude (i.e., price volatility) prior to option expira-tion. In this sense, a straddle trade is non-directional since it canprofit in both bull and bear moves.

Price volatility can be described by several common technical in-dicators including average-true-range (ATR), average-directional in-dex (ADX), standard deviation, and statistical volatility (also calledhistorical volatility). Volatility has been observed to be “mean-revert-ing.” Periods of abnormally high or low short-term price volatilityare followed by price volatility that is closer to the long-term pricevolatility of the underlying asset.(1,3) A short-term drop in price vola-tility (volatility implosion) can be reliably expected to be followed bya sudden volatility increase (volatility explosion). Connors, et. al. haveshown that multiple days of short-term volatility implosion is a pre-dictor of a strong price move.(1,2)

The volatility implosion does not predict the direction of the im-pending price move, but only that there is a high probability that theunderlying asset is going to move away from its current price and bya significant amount. In a previous study(4) we showed that a simple-straddle, options-based strategy designed to exploit a sudden implo-sion of a stock’s volatility, combined with a simple money-manage-ment strategy, with time as the only existing criteria produced supe-rior returns and, therefore, could be used as the basis to developtrading strategies capable of producing superior returns without theneed to correctly predict the direction of a given stock, commodityor market being traded. However, the volatility implosion does notpredict when (how quickly) the explosion price move will develop.Further analysis of the previous study indicated that volatility can re-

EXPLOITING VOLATILITY TO ACHIEVE A TRADING EDGE USINGAN AVERAGE-TRUE RANGE (ATR) SECOND FILTER:

Market-Neutral/Delta-Neutral Trading Using the PRISMTrading Systems

Jeff Morton, M.D., CMT and Randi Schea, M.D.

10


main low and even continue to decrease for several weeks. Thus, notinfrequently, the trade was stopped-out just before the volatility ex-ploded. It appeared that we were successful at defining periods oflow volatility but needed a way of better predicting when the periodof low volatility was ending. In this study we investigated the use of asecond entry filter based on the average-true-range (ATR) of the re-cent daily prices as a way for trying to better define the “end of thelow period of volatility” and thereby improve the overall returns ob-tained using the basic option straddle strategy. At PRISM TradingAdvisors, Inc., this strategy has been successfully implemented to gen-erate superior returns at lower risk than traditional investment port-folio benchmarks.

METHODS AND MATERIALS

System 1 (Time-Based Strategy): This strategy was tested fromNovember 1, 1993 through May 31, 1998 using the stocks that makeup the Dow Jones Industrial Average as a representative sample ofthe broader market. They were chosen because they are a well-knowngroup of stocks that have been designed to represent the market atlarge. Volatility is defined by the price statistical volatility formula:

s.v. = s.d.{log(c/c[1]),n} * square-root (365); where:s.v. = the statistical volatility.s.d. = the standard deviation.c = the closing price of the stock on that day.c[1] = the closing price of the stock of the previous day.

Statistical (or historical) price volatility can be descriptively de-fined as the standard deviation of day-to-day price change using alog-normal distribution and stated as an annualized percentage.Detailed information on statistical volatility is available from the ref-erences.(1,2,3)

Another measure of volatility is the average-true-range (ATR). Itis the average of the true-range (TR) of the daily prices over a speci-fied period of time. True-range (TR) is defined as the greater of:■ the difference between today’s high and today’s low,■ the difference between yesterday’s close and today’s high,■ the difference between yesterday’s close and today’s low.The rules to initiate a trade were as follows:■ Rule 1: 6-day s.v. is 50% or less than the 90-day s.v.■ Rule 2: 10-day s.v. is 50% or less than the 90-day s.v.■ Rule 3: Both Rule #1 and Rule #2 must be satisfied to trigger

completion of the first trade signal.Thus in this study, the first signal, a volatility implosion, occurred

when the 6-day and 10-day historical volatilities were 50% or less thanthe 90-day historical volatility.■ Rule 4: Rule #3 must be satisfied before proceeding to Rule #5.■ Rule 5: The trend of the 14-day ATR, when the first trade signal is

triggered, must be flat or in a downtrend.■ Rule 6: The 14-day (ATR) today must be greater than the 14-day

ATR yesterday.■ Rule 7: The 14-day (ATR) yesterday must be greater than the 14-

day ATR two days ago to initiate the trade.Thus the second signal, the beginning of an increase in volatility,

occurred when the 14-day ATR increased for two consecutive days.When these conditions were met, a signal to initiate a straddle

position was taken the following trading day. The Black-Scholes modelwas used to calculate the options prices that were used to establishthe straddle positions. The opening price of the stock, the actual

implied volatility, and the yield of the 90-day U.S. Treasury Bill wereused to calculate the price of the options. The professional softwarepackage, OpVue 5 version 1.12 (OpVue Systems International), wasused to calculate the options prices assuming a moderate amount ofslippage. For the purposes of this analysis, it was assumed that eachtrade was equally weighted and that an equal dollar amount was in-vested in each trade. Based on the closing stock price, the value ofthe option straddle positions were then calculated using the samemethod described above after 2-weeks, 4-weeks, and 6-weeks respec-tively. Any trading signals generated in a stock with a current openoption straddle position before the end of the 6-week open tradeperiod were ignored. To minimize the effect of time decay and vola-tility, options with greater than 75 days to expiration were used toestablish the straddle positions. The positions were closed out at theend of the 6-week time period with more than 30 days left until expi-ration. To further minimize the effect of volatility, options were pur-chased “at or near the money.” Given the current large variability ofbrokerage fees, the results were calculated without deducting com-mission costs.

System 2 (Money-Management Strategy): As in the original study,a second trading strategy was explored. It was identical to the firsttrading strategy above except that a set of simple money manage-ment rules were added. The rules were designed to 1) cut lossesshort, 2) allow profits to run, and 3) lock in profits.■ Rule #1: A position was closed immediately if a 10% loss occurred.■ Rule #2: If a 5% profit (or greater) was generated, then a trailing

stop of one-half (50%) of the maximum open profit achieved bythe position was placed and the position closed if the 50% trailingstop was violated.

■ Rule #3: If neither Rule #1 or #2 was violated then the positionwas closed out after either four weeks or six weeks.

RESULTS

System 1 (Time-Based Strategy): A total of 230 trades were gener-ated between 11/1/93 and 5/30/98. Numerous parameters of the230 trades were analyzed. The results are summarized in Table 1.The 2-week, 4-week, and 6-week cumulative returns were +88%,+255%, and -151% respectively and are shown in Chart 1. The re-turn of the DJIA over the same time period was +242% (3680.59 to8899.95). The maximum drawdowns for the 2-week, 4-week, and 6-week series were, -143%, (8/24/94 – 3/10/94), -249% (9/1/94 – 3/13/94), and -640% (12/29/93 – 1/26/95). The maximum draw-upsfor the 2-week, 4-week, and 6-week series were, +312% (3/10/94 –11/25/95), +557% (3/13/94 – 1/30/96), and +561% (1/26/95 – 4/24/96). The results of the current study are compared with the re-sults of the prior study in Charts 3 and 4.

System 2 (Money-Management Strategy): A total of 230 tradeswere generated between 11/1/93 and 5/30/98. Numerous param-eters of the 230 trades were analyzed. The results are summarized inTable 2. The 2-week, 4-week, and 6-week cumulative returns were+251%, +979%, and +1470% respectively and are shown in Chart 2.The return of the DJIA over the same time period was +242% (3680.59to 8899.95). The maximum drawdowns for the 2-week, 4-week and 6-week series were -132% (8/24/94 – 3/10/95), -112%, (9/1/94 – 1/19/95) and -125% (9/1/94 – 12/30/94). The maximum draw-upsfor the 2-week, 4-week and 6-week series were +223% (3/10/95 – 11/24/95), +666% (2/7/95 – 4/24/95) and +939% (12/20/94 – 4/24/96). The results of the current study are compared with the results ofthe prior study in Charts 5 and 6.


DISCUSSION

It has been observed that short-term volatility will have a tendencyto revert back to its longer-term mean.(1,3) Connors et.al.(1) have pub-lished the Connors-Hayward Historical Volatility System and showedthat when the ratio of the 10-day versus the 100-day historical volatili-ties was 0.5 or less, there was a tendency for strong stock price movesto follow.

In our previous study at PRISM Trading Advisors, Inc., we con-firmed the phenomenon of volatility-mean reversion by presentingthe first large-scale, option-based analysis while maintaining a strictmarket-neutral/delta-neutral (gamma positive) trading program(4).We showed that a significant price move occurred 75% of the timefollowing a short-term volatility implosion (as defined in the Meth-ods and Materials section).

In the previous study, we chose a relatively straightforward strat-egy of purchasing a straddle. A straddle is the proper balance of putand call options that produce a trade with no directional bias. Astraddle is said to be “delta neutral” and will generate the same profitwhether the underlying asset’s price moves higher or lower. As theasset price moves away from its initial price, one option will increasein value while the other opposing option will decrease in value. Aprofit is generated because the option that is increasing in value willincrease in value at a faster rate than the opposing option is decreas-ing in value. The straddle is said to be “gamma positive” in bothdirections, because one is both long the call option and long the putoption.

This option strategy has a defined maximum risk that is known atthe initiation of the trade. This maximum risk of loss is limited tothe initial purchase costs of the straddle (premium costs of both putand call options). There is no margin call with this straddle strategy.There is an additional way that this strategy can profit. Because theoptions are purchased at the time there has been an acute rapid de-crease in volatility, one should theoretically be purchasing “under-valued” options. As the price of the asset subsequently experiences asharp price move, there will be an associated increase in volatilitywhich will increase the value of all the options that make up thestraddle position. The side of the straddle which is increasing invalue will increase at an even faster rate, while the opposite side ofthe straddle which is decreasing in value will decrease in value at aslower rate. So as to not further complicate the analysis, the exitstrategy for the first system (time-based strategy) for this study waseven more basic; using a time-stop exit criteria.

In the previous study we showed that a 4-week exit produced apositive return over the study period (335%). However, the draw-downs precluded its use as a stand-alone system for real-time trading(-451%). In that study, a simple set of money management ruleswere added to the original system tested. These money managementrules were designed to close-out non-performing trades early beforethey could turn into large losses and kept performing positions openas long as they continued to generate profits. These goals were ac-complished by closing out any position if its value decreased to 90%of it’s initial value (i.e. a 10% loss). A position with open profits hada 50% trailing stop of the maximum open profit achieved by the po-sition at any time open profits exceeded 5%. If neither of these twoconditions occurred, the position was closed-out at the end of sixweeks. As predicted, the 6-week money-management strategy pro-duced both a significantly greater total return (1189%) with a signifi-cantly smaller drawdown (-246%) than the 4-week non-money-man-agement strategy. By closing positions when a loss of 10% had oc-curred, we were able to significantly decrease the amount of losses.

Further analysis of the previous study indicated that volatility can

remain low and even continue to decrease for several weeks. Thus,not infrequently the trade was stopped out just before the subsequentand anticipated volatility exploded. It appeared that we were suc-cessful at defining periods of low volatility but needed a way of betterpredicting when the period of low volatility was ending. It has alsobeen shown that unusually high volatility generally indicates that asustainable trend is underway. Price-range expansion, after a periodof unusually low volatility, indicates that a new sustainable trend isbeginning. In this study, we investigated the use of a second entryfilter based on the average-true-range (ATR) of the recent daily pricesas a way for trying to better define the “end of the low period ofvolatility” and the “beginning of a reversion of volatility back towardsits mean.” It was hoped that the ATR filter would improve the overallreturns obtained using the basic-option-straddle strategy, while simul-taneously decreasing the drawdowns experienced by the original study.

In the original study, the 4-week time-stop yielded the best resultssince it was felt that the 2-week time-stop did not allow for sufficienttime for the anticipated price move to fully develop; total return of+335% versus -191%. The 6-week time-stop allowed for the adverseeffects of time-decay, volatility, and price regression back towards thestock’s initial starting price that eroded the value of the straddle po-sition; total return of +335% versus -84%. Since the addition of thesecond ATR filter was intended to delay the entry into the trade untilthe period of low volatility had ended, we were concerned that theuse of the same 4-week time-stop might experience some of the sameproblems encountered in the original study with the 6-week time-stop. This is in fact what was seen; the 2-week time-stop producedsignificantly better results while the 4-week time-stop produced slightlyworse results when compared to the original study. If one comparesthe time-based systems from the original study with the current study(Table 3), one finds that for the 2-week time-stop the total return (-191% vs. +88%), the average return-per-trade (-0.60% vs. +0.38%),and the maximum drawdown (-424% vs. -143%) were significantlyimproved. For the 4-week time-stop (Table 4), the total return (+335%vs. +255%) and the average return-per-trade (+1.20% vs. +1.11%) wereslightly worse. The maximum drawdown again, however, was signifi-cantly improved (-451% vs. -249%). It was, therefore, concluded thatthe addition of the second ATR filter had a significant positive effecton the original trading system.

As in the original study, the application of a simple set of money-management rules was again explored in the study. Once again, asin the original study, the money-management rules dramatically im-proved the overall returns while simultaneously decreasing the draw-down experienced in the time-based strategy. The combination ofthe money-management rules with the addition of the ATR filter fur-ther improved the results (Tables 5 and 6). For the 6-week time-stopplus money management, there was a 24% improvement in total re-turns (1187% vs. 1470%), a 50% improvement in average return-per-trade (4.25% vs. 6.39%), and a 51% reduction in the magnitude ofthe maximum drawdown (-246% vs. -125%).

The current study also continues to suffer from several limitations.Although moderate slippage was used in all the calculations, the ro-bustness of this study might have been improved if access to real-timestock option bid-ask prices were available for all of the trades investi-gated. Unfortunately, such a large, detailed database is not readilyavailable. Given that the real-time bid-ask prices were not available,the use of the Black-Scholes formula with the known historical in-puts (stock price, implied volatility, 90-day T-bill yield) is an accept-able alternative thereby minimizing any pricing differences betweenthe actual and theoretical option prices systematically throughoutthe time period used in the study.

The current study revealed that the addition of a second ATR fil-


ter to a simple straddle options-based strategy designed to exploit asudden implosion of a stock’s volatility with time as the only existingcriteria, yielded superior results when compared to the same strategywithout the second ATR filter. Although improved, this strategy withthe second ATR filter continues to produce drawdowns that precludeit as a viable trading strategy. The addition of some simple money-management rules dramatically improved the overall returns whilesimultaneously decreasing the excessive drawdowns that plagued theoriginal trading strategy, thereby transforming it into an applicabletrading system for everyday use. This volatility-based, delta-neutralstrategy also is independent of market direction. A market-neutralstrategy and portfolio may be considered as a separate asset class byportfolio managers in the efficient allocation of their investmentportfolios to boost returns while simultaneously decreasing their riskexposure.

In conclusion, this is the second large-scale, trading-research studyto be shared with the trading public that clearly demonstrates howthe phenomenon of price-volatility, mean-reversion can be exploitedby using an options-based, delta-neutral approach. By adding a sec-ond filter based on ATR to signal the end of the “low volatility pe-riod,” the results were significantly improved. Price, time and volatil-ity factors using options-based strategies to further maximize positiveexpectancy continue to represent active areas of real-time tradingresearch at PRISM Trading Advisors, Inc. These results will be thesubject of future articles.

Table 1

System 1 (Time-Based System)2-Week 4-Week 6-Week

Total Return +88% +255% -151%

Average Return per Trade +0.38% +1.11% -0.65%

Maximum Draw-Up +312% +557% +561%

Maximum Draw-Down -143% -249% -640%

Total # Winning Trades 84 86 100

Total # Losing Trades 146 144 130

Max. # of Consecutive Wins 5 4 6

Max. # of Consecutive Loses 11 9 23

Greatest Gain in One Trade +85% +105% +112%

Greatest Loss in One Trade -31% -33% -54%

Chart 1: System 1 (Time-Based System)

Table 2

System 2 (Money-Management System)2-Week 4-Week 6-Week

Total Return +251% +979% +1,470%

Average Return per Trade +1.09% +4.26% +6.39%

Maximum Draw-Up +223% +666% +939%

Maximum Draw-Down -132% -112% -125%

Total # Winning Trades 84 100 105

Total # Losing Trades 146 130 125

Max. # of Consecutive Wins 5 6 6

Max. # of Consecutive Loses 11 9 7

Greatest Gain in One Trade +85% +105% +112%

Greatest Loss in One Trade -10% -10% -10%

Chart 2: System 2 (Money-Managment System)

Table 3

Comparison of Time-Based Systems 2-Week without ATR Filter 2-Week with ATR Filter

Total Return -191% +88%

Average Return per Trade -0.69% +0.38%

Maximum Draw-Up +374% +312%

Maximum Draw-Down -424% -143%

Total # Winning Trades 91 84

Total # Losing Trades 189 146

Max. # of Consecutive Wins 7 5

Max. # of Consecutive Loses 14 11

Greatest Gain in One Trade +88% +85%

Greatest Loss in One Trade -48% -31%

Table 4

Comparison of Time-Based Systems 4-Week without ATR Filter 4-Week with ATR Filter

Total Return +335% +255%

Average Return per Trade +1.20% +1.11%










Table 5

Comparison of Money Management Systems4-week without ATR Filter 4-week with ATR Filter

Total Return +993% +979%



Maximum Draw-Down -188% 112%







Table 6

Comparison of Money Management Systems6-week without ATR Filter 6-week with ATR Filter

Total Return +1189% +1,470%










REFERENCES

1. Connors, L. A., and Hayward, B.E., “Investment Secrets of a HedgeFund Manager,” Probus Publishing, 1995.

2. Connors, L. A: “Professional Traders Journal.” Oceanview Finan-cial Research, Malibu, CA. March 1996, Volume 1, Issue 1.

3. Natenberg, S., “Option Volatility and Pricing. Advanced TradingStrategies and Techniques,” McGraw Hill, 1994.

4. Morton, J. D.: “Exploiting Volatility to Achieve a Trading Edge:Market-Neutral/Delta-Neutral Trading Using the PRISM Trading Sys-tems.” The MTA Journal, Issue 54, pp. 9-12, 2000.


Notes

Notes

NotesNotes

Notes

Notes

Note

s

NOTES

journal of technical analysis (jota). issue 56 (2001, winter)

Economy & Finance

mta journal editors

cmt kirkpatrick

mta journal fallwinter

cmt dorsey

mta member

cmt jeffrey morton

cmt lake quivira

cmt aerodynamic investments