price discovery in cash& futures market-laxmi sn-0483

8/7/2019 Price Discovery in Cash& Futures Market-Laxmi SN-0483

1/73

Price discovery in Cash and Futures Market

M P Birla Institute Of Management 1

A RESEARCH PROJECT

On

Price Discovery in Cash and Future markets

Submitted in partial fulfillment of the requirement for MBA

Degree of Bangalore University

BY

Lakshmi S N

Registration Number

04XQCM6046

Under the guidance of

Dr. N.S.Mallvalli

M.P.Birla Institute of Management

Associate Bharatiya Vidya Bhavan

Bangalore-560001

2004-2006


2/73



DECLARATION

I hereby declare that the research project titled Price discovery in cash and futures

markets is prepared under the guidance of Dr.N.S. Mallavalli in partial

fulfillment of MBA degree of Bangalore University, and is my original work.

This project does not form a part of any report submitted for degree or diploma

under Bangalore University or any other university.

Date :

Place: Bangalore Lakshmi S N


3/73



PRINCIPALS CERTIFICATE

This is to certify that Ms. Lakshmi S N, bearing Registration No: 04XQCM6046

has done a research project on Price Discovery in Cash and Futures Markets

under the guidance of Dr. N.S.Mallavalli, M P Birla Institute of Management,

Bangalore. This has not formed a basis for the award of any degree/diploma for

any other university.

Place: Bangalore Dr.N.S.MALLAVALLI

Date: MPBIM, Bangalore


4/73



GUIDES CERTIFICATE

I hereby declare that the research work embodied in this dissertation entitled

Price discovery in Cash and Futures Markets has been undertaken and

completed by Miss. Lakshmi .S. N under my guidance and supervision.

I also certify that she has fulfilled all the requirements under the covenant

governing the submission of dissertation to the Bangalore University for the award

of MBA Degree.

Place: Bangalore Dr. N S MALLVALLI


5/73



Date: MPBIM, Bangalore

ACKNOWLEDGEMENT

I am thankful to Dr.N.S.Malavalli, Principal, M.P.Birla institute of management,

Bangalore, who has given his valuable support during my project.

I am extremely thankful to Prof. Dr.N.S.Malavalli , M.P.Birla institute of

Management, Bangalore, who has guided me to do this project by giving valuable

suggestions and advice.

My special thanks to Dr. T.V.N Rao and Prof. Santhanam, who provided me the

timely advice and and has helped remarkably to complete the project.

Special thanks to my friend Megha N Bias who made this report reality.

Finally, I express my sincere gratitude to all my friends and well wishers who

helped me to do this project.

LAKSHMI S N


6/73



TABLE OF CONTENTS

CHAPTERS PARTICULARS PAGE NO.

ABSTARCT

1. INTRODUCTION

1.1 Introduction and Background of the study 1 - 7

2. REVIEW OF LITERATURE

Theoretical Framework 8 -15

Past studies 16 - 20

3. RESEARCH METHODOLOGY

3.1 Hypothesis Statements 21

3.2 Data sampling details 22 - 23

3.2 Statistical Models 24 - 29

4. DATA ANALYSIS AND INTERPRETATION

Empirical Results 30 - 38

5. INFERENCES AND CONCLUSIONS 39 - 40

BIBLIOGRAPHY 41

ANNEXURES 42 - 58


7/73



ABSTRACT

One of the important function of Futures market is Price discovery. Futures marketsprovide a mechanism through which information about current and futures Spot prices can be

assimilated and disseminated to all participants in the economy. The ability of future markets to

provide information about price is a central theme for the existence of these markets.

The present study has been done to examine whether the futures market is performing its

primary role of price discovery. The main objectives of the study is to examine whether there

exists any long run equilibrium relationship between spot market and future market and to

identify the lead lag relationship between these markets.

The daily NSE closing prices of Nifty Spot Index and Nifty Index Futures of near month

contracts from 12th

June 2000 to 31st

Mar 2006 have been taken for the study purpose. After

obtaining the statonarity of the series by conducting Augmented Dickey Fuller Test, Granger s

Co integration Test and Johansen Co integration Test have been conducted on the Nifty Spot and

Nifty futures. The empirical results indicate that there exists long run equilibrium between these

two markets. Further Grangers Causality Test results indicate that there exists lead lag

relationship between these Spot market and Future market. It also shows that there is much

feedback from Futures market to Spot market in impounding information in its prices

Finally, from overall empirical results conclude that Futures markets are more efficient in

discovering the Future spot price.


8/73




9/73



INTRODUCTION

One of the important functions of the futures market is price discovery and hedging. Futures

market provides a mechanism through which information about current and future spot can be

assimilated and disseminated to all participants in the economy. The futures trading in India

started with the introduction of index futures on NSE and BSE in June 2000

Futures markets have 2 important functions in the economy providing hedge facilities and price

discovery. The ability of futures markets to provide information about an optimum allocation of

resources in an economy, prices must accurately reflect relative production costs and relative

consumption utilities. Futures markets, by providing a mechanism through which information

about current and futures spot prices can be assimilated and disseminated to all participants in

the economy, help to achieve this goal

All these prices are the result of open and competitive trading and reflect the underlying supply

and demand for a financial asset or a good, both in the present and at various times in the future.

Most specifically, in the case of future delivery, future prices reflect current expectations about

what the supply and demand for a commodity or a financial asset are likely to be at different

times in future. As such future prices discover expected spot prices. The equilibrium spot prices

that are expected to prevail at various times in the future.

In India, derivatives were introduced as a part of financial market reforms to hedge price risk

which started in 1990s. These reforms were aimed at enhancing competition, transparency and

efficiency in the Indian financial market. This was initiated by the Govt. Of India through

L.C.Gupta committee report. The L.C.Gupta committee on derivatives had recommended in1997 its introduction in a phased manner. Accordingly stock index futures were introduced on

BSE and NSE in the first. BSE was the first stock exchange in the country, which commenced

trading in index futures based on the BSE Sensex on June 9, 2000. Immediately, there after, on

June 12, 2000, NSE introduced its trading based on S&P CNX NIFTY. Subsequently, later on,

other products like stock futures on individual securities were introduced in November 2001.


10/73



This was followed by approval of trading in index options based on these two indices and

options on individual securities.

Nifty futures traded on NSE are based on S&P Nifty index. At any point of time, contracts with

one month, two months and three months to expiry are available for trading. These contracts

expire on last Thursday of the expiry month and incase of the last Thursday being holiday, the

contracts expire on previous trading day and have a maximum of the expiry of near month

contract. Nifty futures are presently cash settled. The turnover in derivatives segment has

witnessed significant growth in the last 5 years.

The relationship between stock index spot and futures markets is still attracting the attention of

academics, practitioners and regulators due to both the considerable volume of trading in these

contracts and their role during periods of turbulence in financial markets. An important aspect of

this relationship is the nature of the lead-lag relationship in the returns between equivalent assets

traded in different markets or the predictive power of price movements in one market for those in

the other market.

One of the economic functions of futures contracts is price discovery. Price discovery refers tothe use of futures prices for pricing cash market transactions and its significance depends upon

the above mentioned, close relationship between the prices of futures contracts and the

underlying assets. The essence of the price discovery function of futures markets hinges on

whether new information is reflected first in changes of futures prices or in changes of cash

prices.

In other words, price discovery means whether price changes in futures markets lead price

changes in cash markets more often than the reverse. If that is the case, there exists a lead-lag

relationship between the two markets. Therefore, the futures prices may serve as the market s

expectation of a subsequent delivery period cash price. The share of price discovery originating

in the futures markets has important implications for hedgers and arbitrageurs who use these

markets.


11/73



The introduction of stock index futures cause an increase in volatility in the short run while there

is no significant change in volatility in the long-run (Edwards 1988).This is because futures

markets result in uninformed (irrational) speculators trading in both futures and cash markets,

shocking prices in search of short-term gains. Hodgson and Nicholls (1991) quote that increased

market volatility may increase real interest rates and the cost of capital, leading to a reduction in

the value of investments and loss of confidence in the market. In turn, this can lead to a flow of

capital away from equity markets. Secondly, with increased volatility, regulatory bodies may

interfere in markets to enact further regulations. While these regulations are certainly costly and

may or may not reduce stock price volatility. However, another view is that derivative markets

reduce spot volatility; by providing low cost-contingent strategies, enabling investors to

minimize portfolio risk by transferring speculators from spot markets to futures markets. The low

margins, low transaction costs and the standardized contracts and trading conditions attract risk-

taking speculators to futures. Hence, futures have a stabilizing influence as it adds more

informed traders to the cash market, making it more liquid and, therefore, less volatile. It is seen

that increased spot volatility from futures markets may not be undesirable if induced by objective

new information. In general, the quicker and more accurate prices reflect new information, the

more efficient should be the allocation of resources.


12/73



BACKGROUND OF THE STUDY

Future contract is an agreement to buy and sell an asset at a future time for a certain price which

is traded in an organized exchange and the contracts term are standardized by orgainse

exchange. The institutional environment includes a clearing house to guarantee to all trades and

margin system design to protest the financial integrity of the market place. This system allows

the futures market two social benefits: prices discovery and risk transferring through hedging.

Price discovery reveals how fast one markets reflect new information relative to the other, and

how well the two markets are linked. Risk management illustrates how the price risk of a certain

position in the cash market may be reduced through futures market by hedgers. The dual roles of

price discovery and risk transfer provide benefits that cannot be offered in the cash market alone.

The futures markets serve an important function of price discovery. The following concepts help

to identify the theoretical frame work.

CURENT CASH PRICEAssets are commonly traded in cash markets by a network of dealers. In which each dealer gives

his own prices. This price often differs from one another either because dealers operate

indifferent geographical markets or customers are not aware of price differences that exist. To

get the lowest price, a buyer need a conduct time consuming and costly search I the market.

Futures market provides highly visible prices against which the current cash prices of dealers can

be compared. if prices of futures contract for immediate or near term delivery differ from

dealers cash prices, traders can arbitrage this difference. All cash prices will therefore reflect

accurately the supply and demand equilibrium brought by futures market.


13/73



EXPECTED CASH PRICESince futures contracts are traded for delivery at various points of time future, they reflect current

market expectation about the cash price in future. This is also reflects the market expectations of

the demand and supply of an asset at a particular point of time of the future. If futures prices are

considerably higher than the current cash prices, it reflects the market expectation of relative

shortage of asset in the future, whereas lowest future prices indicate relative surplus of the asset

in the future. Though futures trading, information about the expectations of all the market

participants, commercial hedgers and speculators about the demand and supply for an asset in the

future month are assimilated to produce a single futures price for that month. Futures a market

help to smoother the supply and demand of an asset over a period of time and help to avoid

economic dislocation by the discovery of expected cash prices.

The advent of stock index futures and options has profoundly changed the nature of trading on

stock exchanges. The concern over how trading in futures contracts affects the spot market for

underlying assets has been an interesting subject for investors, market makers, academicians,

exchanges and regulators alike.

These markets offer investors flexibility in altering the composition of their portfolios and in

timing their transactions. Futures markets also provide opportunities to hedge the risks involved

with holding diversified equity portfolios. As a consequence, significant portion of cash market

equity transactions are tied to futures and options market activity. In the Indian context,

derivatives were mainly introduced with a view to curb the increasing volatility of the asset

prices in financial markets; bring about sophisticated risk management tools leading to higher

returns by reducing risk and transaction costs as compared to individual financial assets.

However, it is yet to be known if the introduction of stock index futures has served the purpose

claimed by the regulators. The launch of derivative products has significantly altered the

movement of the share prices in the spot market. The spot and futures market prices are linked

by arbitrage, i.e., participants liquidating positions in one market and taking comparable


14/73



positions at better prices in another market, or choosing to acquire positions in the market with

the most favorable prices. If, for example, the observed futures price is above (below) the

theoretical futures price, arbitrageurs sell (buy) futures and buy (sell) the underlying security,

driving down (up) the price of the futures and driving up (down) the prices of security. This

raises important questions about the effect that index derivatives have on volatility of the spot

market. While there is still disagreement as to whether futures trading increases or decreases the

volatility of spot prices, the question is still an empirical one. However, if one market reacts

faster to information, and the other market is slow to react, a lead-lag relation is observed. The

lead-lag relation between price movements of stock index futures and the underlying cash market

illustrates how fast one market reflects new information relative to the other, and how well the

two markets are linked. Hence, this study attempts to examine the lead lag relationship

between the futures and the underlying spot market.

PROBLEM STATEMENTIt is well known that prices of related securities like prices in spot and futures markets cannot

diverge without bound because they are linked by an arbitrage relationship. The link between this

arbitrage relationship (and the associated cost of carry model) may diverse due to short term

disequilibrium. Therefore the existence of a long run co integration relationship between the spot

and futures prices is the problem to be studied with the Grangers Co integration test and Error

Correction Model.

OBJECTIVES OF THE STUDY To know whether futures trading in India is performing its primary role of price

discovery

To know whether there exists long term equilibrium between futures and spot

prices along with short term equilibrium dynamics.

To know whether there exists lead- lag relationship between spot and futures

market.


15/73



SCOPE OF THE STUDYThe research examines the effectiveness of stock index futures by analyzing the dynamic

interactions and causal relationship between speculative type activity and spot market volatility.

Evidence for bi-directional information flow between speculative activity and volatility are

examined between S&P CNX nifty and S&P CNX Nifty Futures Market. It appears that

investors speculate in the futures market, in particular when faced with volatility in the cash

market. The fluctuations as a result of speculative activity are decreasing over a period of time,

possibly due to the hedging activities taking place in the market. The dynamic interactions

between speculative activity and spot volatility show that index futures are having a stabilizing

effect on underlying spot market.

LIMITATIONS OF THE STUDY The study is limited only to S&P CNX NIFTY and S&P CNX NIFTY Futures.

The study is not done on stocks and stocks futures. Therefore the conclusions

are based on S&P CNX NIFTY and S&P CNX NIFTY Futures.

The study is limited to near month contract. The Mid month contract and Far month

contracts have not been considered.


16/73




17/73



THEORITICAL BACKGROUND

INDEXAn index is a number used to represent the changes in a set of values between a base time period

and the current time. A stock index represents change in the value of a set of stocks which

constitute the index over a base year

INDEX FUTURE

An index future is a derivative whose value is dependant on the value of the underlying asset (eg.

BSE Sensex, S&P CNX NIFTY). While trading on index futures, an investor is basically buying

and selling the basket of securities comprising an index in their relative weights. Unlike

commodity and other futures contracts, Index Future contracts are settled in cash.

FUTURES CONTRACTA futures contract is a standardized, transferable, exchange-traded contract that requires delivery

of a commodity, bond, currency, or stock index, at a specified price, on a specified future date.

Generally, the delivery does not occur; instead, before the contract expires, the holder usually

"squares their position" by paying or receiving the difference between the current market price of

the underlying asset and the price stipulated in the contract.

Unlike options, futures contracts convey an obligation to buy. The risk to the holder is unlimited.Because the payoff pattern is symmetrical, the risk to the seller is unlimited as well. Money lost

and gained by each party on a futures contract are equal and opposite. In other words, a future

trading is a zero-sum proposition.

Futures contracts are forward contracts, meaning they represent a pledge to make a certain


18/73



transaction at a future date. The exchange of assets occurs on the date specified in the contract.

Futures are distinguished from generic forward contracts in that they contain standardized terms,

trade on a formal exchange, are regulated by overseeing agencies, and are guaranteed by

clearinghouses. Also, in order to insure that payment will occur, futures have a margin

requirement that must be settled daily. Finally, by making an offsetting trade, taking delivery of

goods, or arranging for an exchange of goods, futures contracts can be closed.

Trading in futures is regulated by the Securities & Exchange Board of India (SEBI). SEBI exists

to guard against traders controlling the market in an illegal or unethical manner, and to prevent

fraud in the futures market.

ADVANTAGES OF FUTURES TRADING

There are many inherent advantages of trading futures over other investment alternatives such as

savings accounts, stocks, bonds, options, real estate and collectibles.

1. HIGH LEVERAGE..The primary attraction, of course, is the potential for large profits in a short period of time. The

reason that futures trading can be so profitable is the high leverage. To own a futures contract

an investor only has to put up a small fraction of the value of the contract (usually around 10-

20%) as margin. In other words, the investor can trade a much larger amount of the security

than if he bought it outright, so if he has predicted the market movement correctly, his profits

will be multiplied (ten-fold on a 10% deposit). This is an excellent return compared to buying

and taking physical delivery in stocks.


19/73



2. PROFIT IN BOTH BULL & BEAR MARKETS

In futures trading, it is as easy to sell (also referred to as going short) as it is to buy (also referredto as going long). By choosing correctly, you can make money whether prices go up or down.

Therefore, trading in the futures markets offers the opportunity to profit from any potential

economic scenario. Regardless of whether we have inflation or deflation, boom or depression,

hurricanes, droughts, famines or freezes, there is always the potential for profit making

opportunities.

3. LOWER TRANSACTION COSTAnother advantage of futures trading is much lower relative commissions. Your commission for

trading a futures contract is one tenth of a percent (0.10-0.20%). Commissions on individual

stocks are typically as much as one percent for both buying and selling.

4. HIGH LIQUIDITY..Most futures markets are very liquid, i.e. there are huge amounts of contracts traded every day.

This ensures that market orders can be placed very quickly as there are always buyers and sellers

for most contracts.

S & P CNX NIFTY

S&P CNX Nifty is a well diversified 50 stock index accounting for 25 sectors of the economy. It

is used for a variety of purposes such as benchmarking fund portfolios, index based derivatives

and index funds.

S&P CNX Nifty is owned and managed by India Index Services and Products Ltd. (IISL), which

is a joint venture between NSE and CRISIL.


20/73



The average total traded value for the last six months of all Nifty stocks is approximately 49.8%

of the traded value of all stocks on the NSE

Nifty stocks represent about 56.5% of the total market capitalization as on March

31, 2006.

Impact cost of the S&P CNX Nifty for a portfolio size of Rs.5 million is 0.07%

S&P CNX Nifty is professionally maintained and is ideal for derivatives trading

S&P CNX NIFTY FUTURESA futures contract is a forward contract, which is traded on an Exchange. NSE commenced

trading in index futures on June 12, 2000. The index futures contracts are based on the popular

market benchmark S&P CNX NIFTY index.

NSE defines the characteristics of the futures contract such as the underlying index, market lot,

and the maturity date of the contract. The futures contracts are available for trading from

introduction to the expiry date.

CONTRACT SPECIFICATIONSThe security descriptor for the S&P CNX Nifty futures contracts is:

Market type: N

Instrument Type: FUTIDX

Underlying: NIFTY

Expiry date: Date of contract expiry

Instrument type represents the instrument i.e. Futures on Index.

Underlying symbol denotes the underlying index which is S&P CNX Nifty


21/73



Expiry date identifies the date of expiry of the contract

TRADING CYCLES&P CNX Nifty futures contracts have a maximum of 3-month trading cycle - the near month

(one), the next month (two) and the far month (three). A new contract is introduced on the

trading day following the expiry of the near month contract. The new contract will be introduced

for three month duration. This way, at any point in time, there will be 3 contracts available for

trading in the market i.e., one near month, one mid month and one far month duration

respectively.

EXPIRYDAYS&P CNX Nifty futures contracts expire on the last Thursday of the expiry month. If the last

Thursday is a trading holiday, the contracts expire on the previous trading day.

FUTURES PRICE FORMATIONThere are two views in futures literature regarding price formation process of futures prices. In

the first, the inter-temporal relationship between cash and future prices of futures is explained by

cost of carry model.

The theory of inter-temporal relationship between cash and futures prices can be explained with

the cost of carry model. This is given as

Ft = St e r t

r = holding costs

t = time till expiration.

If F > St er t

or F < St er t

then futures price is away from its fair value and arbitrage opportunities


22/73



exist there. The relationship between the cash market and futures market are thought of as being

maintained by professional arbitragers who employ computerized trading systems to capitalize

on deviations of stock index futures prices from perceived fair values or whenever parity

represented by equation (1) is violated. If actual futures price exceed their perceived fundamental

value, the futures contract is overvalued. In this situation, arbitragers simultaneously take

positions in the underlying market and futures market, and hence lock in a secure payoff. When

the futures contract is overvalued arbitragers sell futures contract and buy stocks in the

underlying market while on the other hand, an undervalued futures contract triggers a short

arbitrage position where arbitragers buy futures and sell stocks simultaneously in both the

markets.

If F0 = E (St) , the futures price will drift up or down only if the market changes its views about

the expected future spot price. Over a long period of time, it assumed the market revises its

expectations about futures spot prices upward as often as it does so downward. It follows that

when F0 = E (St), the average profit from holding futures contracts over a long period of time

should be zero. The F0 < E (St) situation corresponds to the positive systematic risk situation.

Because the futures price and the spot price must be equal at maturity of the futures contract, this

implies that that a futures price should , on average , drift up and a trader should over a long

period of time make positive profits from consistently holding futures positions. Similarly , theF0 > E(ST) situation implies that a trader should over a long period of time make positive profits

from consistently holding short futures positions.

The second view splits the futures price into an expected risk premium and a forecast of a future

spot price. The basis can be expressed as a sum of an expected premium and an expected change

in the spot price.

FtTSt = Et (p(t,T)+ Et (STSt)

Here the expected premium Et = (P (t, T) is defined as the bias of future price as a forecast of the

future spot price.


23/73



Et [p(t,T)] = FtT

- Et[ST]

Fama and French argue that the theory of holding cost, equation(1) and equation(2) are

alternative but not competitive views of the basis, and the variation in expected change in the

spot price in equation(2) translates into variation in the interest rate and stable relationship

between spot and futures prices. In addition, for the future price to be an unbiased predictor of

eventual spot price premium must be zero, the future price should lead spot price.

The implication of the cost of carry model is that in perfectly efficient and continuous spot and

futures market, price adjustments are instantaneous. The observed relationship between price

changes in the two markets will be noisy due to market imperfections and because price

observations from the two markets are not simultaneous. Therefore the normal relationship

between stock index and stock index futures would be bounded by a lower and upper bound of

no arbitrage trading band. This no arbitrage trading band is determined by market imperfections.

Besides this, a lead-lag relation between price changes in the two markets is very likely, if there

are economic incentives for traders to use one market over the other.

DEVIATIONS OF FUTURES PRICES FROM THEIR FAIR VALUESThe trading strategies most closely identified with the term "stock index arbitrage" are those that

profit from deviations of a stock index futures price from its fair value. A short NIFTY futures

contract effectively converts an investment in NIFTY portfolio into a short term loan at a risk-

free rate of interest. Similarly, a long NIFTY futures contract converts a short position in the

NIFTY portfolio into a short-term debt at a risk-free rate of interest.

This rate of interest is not explicitly quoted but is a synthetic rate derived from the relationship

between the futures and cash index prices. An investor who buys the NIFTY portfolio spot and

simultaneously sells it forward with a short NIFTY futures position is nevertheless "lending" at

this implied rate, just as he would lend at money market rate. Similarly, an investor who shorts

the NIFTY portfolio and simultaneously buys the portfolio forward with long NIFTY futures


24/73



position is "borrowing" at the rate of interest implied by the futures price "implied futures rate".

Since an investor can lend or borrow at the implied futures rate, a profitable arbitrage

opportunity arises whenever the implied futures rate differs from the money market rate at which

the investor can lend or borrow. If the implied futures rate is greater than an investor's borrowing

rate, the futures price is deemed "rich", i.e., it is greater than its fair value. To profit from the

mispriced futures, the investor borrows at the money market rate and lends at the higher implied

futures rate.

Conversely, the futures price is deemed "cheap", i.e., smaller than its fair value, if the implied

futures rate of interest is smaller than the investor's lending rate. In this case, the investor

borrows at the implied futures rate and lends at the higher money market rate.


25/73



PAST STUDIES

The relationship between spot and futures market has been the focus of much literature for sometime. This include the fact that futures markets tend to have less trading constraints than the cash

markets, leading to future markets being more informationally efficient as the marginal cost from

trading will be less than in the cash markets. This is further compounded by the fact that futures

market tend to have lower transaction cost and higher liquidity.

There is wide body of research which clearly indicates futures market lead spot market and

suggest that this provides evidence of futures markets are acting as a vehicle for price discovery

within the corresponding spot markets

Pascal Alphonse *1 studied (2000) the information linkage between the Nifty Index and Nifty

Futures and its near-month index futures contract and the role (lead or lag) that the futures

market plays using daily closing futures and cash prices.

The study is concerned with the aggregation of information in the French Stock Index Cash and

Futures Markets. The Results indicate that deviations from the equilibrium relationship linking

cash and futures prices originate from information arrivals in the futures market and theta at least

95 % of their price discovery is achieved in this market.

Arbitrage relationship (cost of carry pricing model) and co integration relationship between the

spot and futures prices has been extensively used. Literature suggests inference concerning spot

and future price dynamics should be based on ERROR CORRECTION MODEL. It suggests

that the spot and futures market should share aCommon

Stochastic Trend Model inspired from

HASBROUCK (1995) . Inference concerning error connection mechanism may be obtained from

JOHANSEN (1988) ECM MODEL .Inference concerning the common (unobserved) efficient

price may be obtained from the vector moving average.

*1

indicates Reference article no 1 ---- Pascal Alphonse----See Bibliography


26/73



Ash Narayan Sah and Anil Kumar *2 (2006) studied whether the derivatives trading are

performing its primary role of price discovery. The main objectives of the study were to know to

know whether there exists feedback mechanism between NIFTY spot and NIFTY futures. Daily

NSE data pertaining to stock index futures taken from JUNE 12 th 2000 to Mar 31st 2005. Stoll

and Whaley 1990, examined intra day price changes from S&P 500 and MM stock index and

futures contracts for serial correlation via-ARMA (p,q) process The Co integration approach and

an ECM using the ENGLE-GRANGER methodology is applied to capture both the long run and

short run dynamics of spot and futures prices. Testing of stationary of series is performed by

using augmented Dickey Fuller Test.

Results indicate that there exists a long run relationship between NIFTY spot and futures price.

Presence of co integration between NIFTY spot and futures indicates market is not efficient. One

can combine information of NIFTY spot and futures prices to predict futures NIFTY spot price.

ECM leads to the conclusion that there exists feedback between NIFTY spot and NIFTY futures.

Results also show spot market leads the futures market and price discovery takes place in both

the markets.

Dimitris F. Keniurgois3

(Oct2004) examined the relationship between price movements of

FTSE/ASE 20, 3 months index and the underlying cash market in Athens Stock Exchange.FTSE/ASE 20 futures market, as the first organized Greek derivatives market, established in Aug

1999. Co integration tests are used and an Error Correction Model is developed in order to

examine the relationship between price movements of FTSE/ASE 20, 3 months index and the

underlying cash market in Athens Stock Exchange. the results show the presence of a bie

directional causality between Stock index spot and futures market , indicting that the newly

established ADEX can provide futures contracts that serve as a focal point of information

assimilation and fulfill their price discovery. The Co integration and ECM is used to examine the

co movements between the series. ADF test is done to know the stationarity of the series.

*2

indicates Reference article no 2 ---- Ash Narayan Sah and Anil Kumar----See Annexure

*3indicates Reference article no 3 ---- Dimitris F. Keniurgois-----See Annexure


27/73



Daily data are used during the period from Aug 1999 to June 2002. The logs of the spot and

futures prices are used. Futures prices are of nearby contracts. The results Engle Granger and

Johansen methods of co integration indicate that the presence of a bi directional causality

between the spot index and futures index markets, thus informational linkage between them. This

empirical evidence suggests that the newly established ADEX market provide futures contracts

that can be used as vehicles for price discovery. The temporal relationship between Chicago Corn

and Soyaben cash prices, nearby futures prices and interest rates is examined using daily 1980

1989 data. Johansen Co integration tests suggest joint movement of the 3 series over the data

period considered.

Hector. O. Zapata and T. Randall Fortenbery *4

(1995) in their study examined the relationship

between Chicago corn, Soyaben cash prices, nearby future prices. The Co integration approach

of Granger is used to test the relationship between cash, futures and interest rates. Error

Correction Model used to overcome the short term disequilibrium. Daily closing prices for corn

and Soyabean futures contract from 1980 89, daily Chicago and Soyabean cash prices for the

period 1980 89 and 90 day US T- bill rates for the same period. Futures contracts observed are

the one closest to maturity.

The empirical tests show that cash, futures and interest rates prices are non stationary Of order

one. This suggest that Chicago cash prices and nearby futures prices behave a stochastic trend

variables during all 10 years and 90 day T bill rate has stochastic trend during most years.

Results support the hypothesis that futures market provide a price discovery function and suggest

that cash market are not merely a derivative market of the futures. The findings of co integration

over the aggregate data period suggest that there is long run causation either from the futures to

the cash market or vice versa. The relationship between the stock markets of the developed

countries has been examined extensively in the literature. The study examines the

interdependence of the 3 major stock markets in South Asia.

*4



28/73



Golaka .C. Nath and Sunil Varma *5 (June 2003) in their study have taken Daily stock market

data from Jan 19994 to Nov 2002. Sock market indices of India (NSE NIFTY), Singapore (STI)

and (TAIEX) is been examined for their study purpose. Dynamic linkage is examined using

Grangers Causality Concept (1969, 1988). ADF is used to test the stationarity of the series.

ECM was used since data series ware nonstationary. Johansen Methodology is been used to

overcome the many deficiencies brought forward by a number of researchers The research has

been done on the 3 important capital markets in Asian region and during the period from 1994 to

2002. The literatures suggest that existence of significance interactions between the various

equity markets, the empirical results shows that generally returns in these markets are not

interrelated and there is no long run equilibrium. Therefore the results suggests that international

investors can achieve long term gains by investing in the stock market as the market under study

have been generally independent.

Stoll and Whaley(1990) considered these two latter factors play vital role in explaining why

future markets tend to exhibit price leadership effects over stock markets. In their study Stoll and

Whaley examined intra-day price changes from S&P 500 and MM Stock index and Futures

contracts for serial correlation via and ARMA (p,q) process. They found strong evidence of

futures markets leading stock indices.

Further Research by Antony and Miller(1988) have suggested investors prefer trading in the

derivatives markets rather that the stock markets because of market frictions such as transaction

costs and capital requirements. Evidence of future price leadership over spot prices is also not

confined to research on the US markets. By examining the lead-lag relationship between stock

index prices and index futures contracts within an ARMA (p, q) frame work, noted futures price

leadership for the German DAX index futures contract. Similar results were also obtained by

Shyy, Vijayaraghavan and Scott-quinn for the French CAC index futures contract. This shows

whether there are lead-lag relationships between spot and futures market.

*5



29/73



In India, Thenmozi and Raju (2002) and Karnade (2003) made earliest studies on the lead-lag

and price discovery of futures market in India. This tries to unravel the dynamic relationship

between the futures and spot market in India and will provide some early empirical evidence in

this regard.

The past studies have done to test the price transmission process have used mainly the regression

analysis. However, if price series are not stationary, a phenomenon typical in financial markets,

then standard statistical tests of parameter restrictions is not reliable (Elam and Dixon, 1988).

Thus, for overcoming the problems of non-stationary price series and due to the fact that price

discovery deals with short-run and long-run departures from a presumed equilibrium relation, the

introduction of co integration analysis with error correction models is fortuitous.


30/73




31/73



HYPOTHESIS STATEMENTS COINTEGARTION TEST

HYPOTHESIS STATEMENT 1Ho: There is no significant co integration between spot and futures markets

H1: There is significant co integration between spot and futures market

CAUSALITY TESTSHYPOTHESIS STATEMENT 2

Ho: Future markets significantly do not lead the spot market

H1: Future markets significantly lead the spot market

HYPOTHESIS STATEMENT 3

Ho: spot market significantly does not lead the futures market

H1: spot markets significantly lead the futures market


32/73



DATA COLLECTIONThe data employed in this study comprises of 1460 daily observations on the NIFTY stock index

and stock index Futures contracts (12-03-2000 to 31-03-2006)

Closing prices of spot indices are obtained from the YAHOOFINANCE website and closing

prices of futures contracts are obtained from official website NSEINDIA (www.nseindia.com).

The NIFTY index comprises of 50 Indian companies, quoted on the NATIONAL STOCK

EXCHANGE (NSE), with the largest market capitalization (blue chips) in the Indian equity

market.

Futures contracts are taken which are quoted on the NATIONAL COMMODITY AND

DERIVATIVE EXCHANGE.

NIFTY futures contracts have a maximum of 3-month trading cycle - the near month (one), the

next month (two) and the far month (three).

We have confined the analysis to the near contract because of high trading volume and greater

liquidity in the F & O market and the preliminary research showed that there was not very

much difference between the closing prices of the nearest and second contract.

Daily data are preferred in this study is confined to 6 years which is helpful to get the betterresults.


33/73



DATA TYPE

The present research makes use of secondary data available in the NSEINDIA website

pertaining to the daily closing prices data of Nifty index and Nifty index futures.

DATA SAMPLEDaily NSE data pertaining to NIFTY index futures and stock futures.

PPEERRIIOODDOOFFSSAAMMPPLLEE

Daily observation from JUNE12th 2000 to March 31st 2006

STATISTICAL MODELS APPLIED Augmented Dickey Fuller test to test the stationary of the series.

Grangers co integration approach and Johansen co integration test for co integration

between the series.

Grangers Causality test.

STATISTICAL SOFTTWARE PACKAGES USED

E viewsThis software has been used to conduct the Augmented Dickey Fuller Unit root Test,

Johansen Co integration Test and Grangers Causality Test

SPSSRegression coefficients for Grangers co integration test are obtained by using this

Software


34/73



METHODOLOGY

In this study, and Johansen cointegration Approach and Engle-Granger methodology is applied to

capture both the long run and the short run dynamics of stock returns and inflation rates. Before

doing co-integration analysis, it is necessary to test whether the time series are stationary at

levels by running Augmented Dickey fuller (ADF) test on the series. Because most time series

are non stationary in levels, and the original data need to be transformed to obtain stationary

series. And then the granger causality test is done to test the causal relationship between stock

returns and inflation.

STATIONARITYAccording to Engle and Granger, a time series is said to be stationary if displacement over time

does not alter the characteristics of a series in a sense that probability distribution remains

constant over time. In other words, the mean, variance and co-variance of the series should be

constant over time. The degree of co-integration is closely related with stationary.

The empirical works based on time series data assumes that the underlying time series is

stationary. In regressing a time series variable on another time series variables, one often obtains

a very high R2 (residuals) even though there is no meaningful relationship between the two

variables. This situation exemplifies the problem of spurious or nonsense regression, which

arises when data is non stationary.

A series is said to be integrated of order one [I (1)] if it has to be differentiated once before

becoming stationary. Similarly, a series is of order two [I(2)] if it has to be differentiated twice

before becoming stationary.


35/73



THEORY OF STATIONARITY

Following are different ways of examining about whether a time series variable Xt is stationary

or has a unit root:

In the AR (1) model, if =1, then X has a unit root. If |< 1 then X is stationary.

If X has a unit root, and then its autocorrelations will be near one and will not drop

much as a lag length increases.

If X has a unit root, and then it will have a long memory. Stationary time series do not

have long memory.

If X has a unit root then the series will exhibit trend behavior.

If X has a unit root, then X will be stationary. For this reason, series with unit root are often

referred to as difference stationary series

This means that if the appropriate order of the AR process is lag2 rather than lag1, the term Yt-

1 should be added to the regression model. A test of whether there is a unit root can be carried

out in the same way as for the DF test, with the test statistics provided by the t statistics of the

coefficient. If= 0 then there is a unit root. The same reasoning can be extended for a generic

AR (p) process.


36/73



TESTING STATIONARITY BY UNIT ROOT TEST OF DICKEY FULLER

HYPOTHESIS STATEMENTH0: Series has Unit root : Non Stationary

H1: Series does not have Unit root : Stationary

Dickey Fuller test involve estimating regression equation and carrying out the hypothesis test.

The AR (1) process is.

Yt = C + Yt-1+ t

Where c and ..are parameters and is to be white noise. If -1 < < 1, then Y is stationary series.

While if = 1, y is non stationary series. Therefore, the hypothesis of a stationary series is

involves whether the absolute value of b is strictly less than one. The test is carried out by

estimating an equation with Yt-1 subtracted from both sides of the equation.

yt = C + t-1 + tWhere and the null and alternative hypotheses are

Ho: = 0 ..Non Stationary

H1: = 0 ..Stationary

The usual t-statistic under the null hypothesis of a unit root does not have the conventional t-

distribution. Dickey and fuller (1979) showed that the distribution under the null hypothesis is

nonstandard and simulated the critical values for selected sample sizes. More recently,

Mackinnon (19991) has implemented a much larger set of simulations than those tabulated by

Dickey and Fuller.


37/73



UNIT ROOT TEST BY AUGMENETD DICKEY FULLER TEST

The simple Unit root test is valid only if the series as an AR (1) Process. If the series is correlated

at high order lags, the assumption of white noise disturbances is violated. The ADF controls for

high- order correlation by adding lagged difference terms of the dependent variable to the right-

hand side of the regression

yt = C + t-1 + 1 yt-1 + 2 y t-2 + ..+ p y t-p + t

This augmented specification is then tested

H0: = 0 Non Stationary

H1: = 0 Stationary

In general, the procedure start with whether the variables X and Y in its level form is stationary.

If the hypothesis is rejected, then the series is transformed into first difference of the variable and

tested for stationarity. If first difference series is stationary, this implies that X and Y are I(1).

GGRRAANNGGEERRSSCCOO--IINNTTEEGGRRAATTIIOONNTTEESSTT

Granger introduced the concept of co-integration when he wrote that two variables may move

together though individually they are non stationary. Co-integration is based on the long run

relationship between variables. The idea arises from considering equilibrium relationships, where

equilibrium is a stationary point characterized by forces that tend to push the variables back

toward equilibrium.

In general, if Yt and Xt are both integrated of order I(d), then any linear combination or the two

series will also be I(d).. That is, the residuals obtained on regressing Yt on Xt are I(d).


38/73



If two or more series are co integrated then even though the series themselves may be non

stationary, they will move closely together over time and their difference will be stationary. Their

long run relationship is the equilibrium to which the system converges overtime and the

disturbance term Et can be construed as the disequilibrium error or the distance that the system is

away from equilibrium at time t

The Engle granger co integration test is a two step process:

1. First estimating an ordinary least square (OLS) regression on the data. A regression of one

integrated variable on the other integrated variables (x on y and y on x).

Yt = a + bx t + e t

X & Y will be co-integrated if and only ifet is stationary.

2. Then testing these residuals from regression equation for stationarity using a unit root test of

ADF.

JOHANSEN CO INTEGRATION TESTJohansens methodology for investigating co integration in a multivariate system has been

preferred by many economists. it employs a better properties than the Engle granger s method

and its less bias when there are more than two variables. This method is based on the Eigen

values and Likelihood ratio statistics. The Johansen tests seek the linear combination which is

most stationary where as the Engle Granger tests being based on OLS seek the linear

combination having minimum variance. It identifies whether the series are co integrated or notby examining the existence of no of co integrating equations in them. It checks the trace statistics

to the null hypothesis of no co integration equations and checks for the alternative hypothesis of

more than one co integrating equations.


39/73



GGRRAANNGGEERRSSCCAAUUSSAALLIITTYYTTEESSTT

Although regression analysis deals with the dependence of one variable on other variables, it

does not necessarily imply causation. In other words, the existence of a relationship between

variables does not prove causality or direction of influence. More generally, since the future

cannot predict the past, if variable X causes variable Y, then changes in X should precede

changes in Y.

Granger causality is a technique for determining whether one time series is useful in forecasting

another. Ordinarily, regressions reflect "mere" correlations, but Clive Granger causality test

shows about the causality between two series.

It measures the significance of past values of variable X in explaining variable Y, taking into

account the effect of past values of variable Y itself. Usually causal relations are tested both

ways, from X to Y and from Y to X.

A time seriesXis said to Granger-cause Yif it can be shown, usually through a series of F-tests

on lagged values ofX (and with lagged values of Y also known), that those X values provide

statistically significant information on future values ofY.

The test works by first doing a regression ofY on lagged values ofY. Once the appropriate

lag interval for Y is proved significant (t-stat or p-value), subsequent regressions for lagged

levels ofX are performed and added to the regression provided that they 1) are significant in of

themselves and 2) add explanatory power to the model. This can be repeated for multiple X's

(with each X being tested independently of other X's, but in conjunction with the proven lag

level ofY). More than 1 lag level of a variable can be included in the final regression model,provided it is statistically significant and provides explanatory power.

If the computed f value exceeds the critical F values at the chosen level of significance, we reject

the null hypothesis, in which case the lagged y belongs in the regression. i.e. Y causes X.The

whole procedure should be repeated to test whether X causes Y.


40/73




41/73



EMPIRICAL RESULTSGGRRAAPPHHSSHHOOWWIINNGGDDAAIILLYYPPRRIICCEEMMOOVVEEMMEENNTTSS

Movement of Daily Closing Prices Of NIFTY Spot

From 12 June 2000 to Mar 31 2006

0

500

1000

1500

2000

2500

3000

3500

4000

No of Observations

S

potclosingPrices

Movement fo futures daily prices from June 12 2000 to

31 Mar 2006

0

500

1000

1500

2000

2500

3000

3500

No of observations

Futureclosingprices

INTERPRETATIONThe above graphs show the daily price movements of Nifty index and Nifty Futures from 12

th

Jun 2000 to 31st

Mar 2006. These movements show that the series are non stationery since they

are showing a upward trend as the time changing. This has been further tested by following


42/73



Augmented dickey fuller Test.

AADDFF UUNNIITTRROOOOTTTTEESSTTRREESSUULLTTSS FFOORR SSTTAATTIIOONNAARRIITTYYOOFF TTHHEE

SSEERRIIEESS

AATTLLEEVVEELLSS

H0 = Unit root : Non Stationary

H1 = No Unit root : Stationary

TTAABBLLEE0011

Series Constraints Test Statistics at 1% at 5% at 10%

None 3.30207 ** -2.5671 -1.9396 -1.6157

Intercept 3.04568 ** -2.5671 -1.9396 -1.6157Nifty spot prices

(At lag 0) Trend and Intercept 0.09926 ** -2.5671 -1.9396 -1.6157

(** indicates acceptance of null hypothesis )

TABLE 02Series Constraints Test Statistics at 1% at 5% at 10%

None 1.174202 ** -2.5671 -1.9396 -1.6157

Intercept 0.293061 ** -2.5671 -1.9396 -1.6157Nifty future prices

(At lag 0) Trend and Intercept 0.293061 ** -2.5671 -1.9396 -1.6157

(** indicates acceptance of null hypothesis )

INTERPRETATIONThe ADF test has been conducted on both the nifty spot and nifty futures time series at their

levels. The test has been done at none, intercept, intercept and trend respectively. The results

show that the null hypothesis is accepted since ADF calculated values for nifty spot and nifty

futures are greater than the critical values at all the levels. (1%, 5%, 10%). There exists unit root

in the series. Therefore the both the series are Non Stationary. Since the results are showing that

the series are Nonstationary, then the series are transformed into first order difference to check


43/73



the Stationarity

** Reference Tables No 1, 2, 3, 4, 5, 6 ------- See Annexure

AT FIRST ORDER DIFFRENCEH0 = Unit root : Non Stationary

H1 = No Unit root : Stationary

TABLE 03Series Constraints Test Statistics Durbin

Watson Test

at 1% at 5% at 10%

None -33.55160 * 1.965572 -2.5671 -1.9396 -1.6157

Intercept -33.65837 * 1.965593 -2.5671 -1.9396 -1.6157Nifty spot

prices

(At lag 0) Trend and Intercept -33.93419 * 1.965648 -2.5671 -1.9396 -1.6157

(* indicates rejection of null hypothesis )

TABLE 04Series Constraints Test Statistics Durbin

Watson Test

at 1% at 5% at 10%

None -53.82697 * 2.106065 -2.5671 -1.9396 -1.6157

Intercept -53.88135 * 2.107857 -2.5671 -1.9396 -1.6157Nifty spot

prices

(At lag 0) Trend and Intercept -54.04371 * 2.112291 -2.5671 -1.9396 -1.6157



44/73



** Reference Tables No 7, 8, 9, 10, 11, 12 ------- See Annexure

INTERPRETATIONThe above Tables 3 and 4 show the results regarding the ADF test statistics after transforming the

series into first order difference .The ADF Test has been conducted to check the unit root in the

series. The ADF test statistics calculated for Nifty spot prices are 33.55160 and -53.82697 for

Nifty futures at none respectively which is not significantly different from the values at Intercept

and Trend & Intercept. These calculated values are less than the Mac Kinnon Critical values at

all the levels (1%, 5%, and 10%). This shows that the null hypotheses of Unit root is rejected at

all the levels. Therefore the both the series are Stationary.

Further, Durbin Watson test has been done to find out the auto correlation in the series. The

calculated values for spot prices and futures prices are 1.965572 and 2.106065 at none

respectively which is not significantly different from the values at Intercept and Trend &

Intercept. These values are close or greater than threshold value of 2. These show that the series

are Stationary at their first order difference.


45/73



RREESSIIDDUUAALLBBAASSEEDDGGAARRNNGGEERRSSCCOOIINNTTEEGGRRAATTIIOONNTTEESSTT

After obtaining the statinarity of the series at their first order difference, next residual based

grangers co integration test has been conducted at their levels to know the long run relationship

between the Nifty spot prices and Nifty Futures prices. The residuals are obtained by following

the grangers Ordinary Least Square method.

Residuals for Spote t = St - - F t ------------------------(1)

Residuals for futurese t = Ft - - S t ----------------------(2)

TABLE 05

In the above equations andare the regression coefficients which are obtained by regressing

the nifty index prices against the nifty futures prices. The regression coefficient is positive for

both the series which gives a some idea that they are positively co integrated. After obtaining the

regression coefficients, the residuals are obtained form the equation (1) and (2). But this

relationship has been further studied by following ADF test.

** Reference Tables No 13, 14 ------- See Annexure

Regression coefficients Spot prices Future Prices

-11.657 16.309

1.012 0.985


46/73



AATT LLEEVVEELL

TABLE 06ADF TESTS TEST

STATISTICS

DURBIN

WATSON

AT 1% AT 5% AT 10%

Spot prices -38.15346 * 1.9999 -2.5671 -1.9396 -1.6157

Future PricesNone Lag 0

-38.15707 * 2.0000 -2.5671 -1.9396 -1.6157


The above Table 6 shows the results regarding the ADF test statistics obtained for residuals

obtained from the Grangers Co integration Test .The ADF Test has been conducted to check the

unit root in the series. The ADF test statistics calculated for Nifty spot prices are -38.15346

-38.15707 for Nifty futures at none respectively. These calculated values are less than the Mac

Kinnon Critical values at all the levels (1%, 5%, and 10%). This shows that the null hypotheses

of Unit root is rejected at all the levels. Therefore the both the series are Stationary which shows

that the series are co integrated.

Further, Durbin Watson test has been done to find out the auto correlation in the series. The

calculated values for spot prices and futures prices are 1.965572 and 2.106065 at none

respectively which is not significantly different from the values at Intercept and Trend &

Intercept. These values are close or greater than threshold value of 2. These show that the series

are Stationary at their first order difference.

** Reference Tables No 15, 16 ------- See Annexure


47/73



JOHANSEN COINTEGRATION TEST

Further Johansen Co integration test has been conducted on the price series which is preferred by

many economists. This test is very much suitable for the multivariate system and moreover this

overcomes the limitations of the Grangers co integration test.

FOR BOTH SPOT AND FUTURE PRICESH0: r = 0 : No co integration

H1: r = 1 : co integration

TABLE 07

INTERPRETATIONThe above table shows the results of co integration between the Nifty spot prices and Nifty

futures prices. This test has been conducted at various lags to check the influence of past value of

one series against another series. The test results reveal that the calculated likelihood ratio isgreater than the critical values at all the lags compared to its critical values. The null hypothesis

is rejected at all the levels (1%, 5%, and 10%). That means three exists more than 1 co

integration equations. This shows that there is long run equilibrium relationship between Nifty

spot prices and Nifty future prices.

** Reference Tables No 17, 18, and 19 ------- See Annexure

TESTS NULL

HYPOTHESIS

EIGEN

VALUES

LIKELI HOOD

RATIO

AT 1% AT 5%

Lag 0 r = 0 0.376356 699.6574* 12.53 16.31

Lag 1 r = 0 0.390663 730.9832* 12.53 16.31

Lag 5 r = 0 0.130991 211.4444* 12.53 16.31

r indicates no of co integration equations



48/73



GGRRAANNGGEERRSSCCAAUUAASSLLIITTYYTTEESSTT

Since there exists long run relationship between Nifty spot prices and nifty futures prices, there

will be causal relationship between these series. That means there exists lead lag relationshipbetween them. This should be examined to know which market lead and which market lag in

impounding the information in its prices. This relationship is examined by conducting Grangers

Causality test.

H0 = Futures does not cause Spot market

H1 = Futures causes Spot market

TTAABBLLEE0088

IINNTTEERRPPRREETTAATTIIOONN

The above table shows the results that the calculated F values are significantly greater from the

critical values at all the lags (2, 5, 12). That means the null hypothesis is rejected and the

alternative hypothesis is accepted .That shows that there is causality from futures to spot market

in impounding the information in its prices. This tells that, any changes in futures prices causes

change in Spot prices. The prices changes in futures prices precede the changes in the Spot

prices. Moreover the probability values are close to 0 which tells that the probability of

occurrence of null hypothesis is 0.

** Reference Tables No 20------- See Annexure

TESTS F- STATISTICS

(FUTERES)

P VALUES AT 1% AT 5% AT 10%

Lag 2 439.112* 0.00000 1 1 1

Lag 5 162.731* 0.00000 1 1 1

Lag 12 67.8706* 0.00000 1 1 1

(* rejection of null hypothesis )

* Reference tables


49/73



TTAABBLLEE0099

INTERPRETATIONThe above table shows the results that the calculated F value is greater from the critical values at

lag 2but its p value is significantly different from zero. Moreover the calculated F values are

significantly lower at 5 and 12 lags which show that the null hypothesis is accepted at 5 and 12

lags. Moreover the probability values are close to 1 which tells that the probability of rejection of

null hypothesis is higher. That shows that there is no much causality from Spot market to Futures

market in impounding the information in its prices. This tells that, any changes in Spot prices do

not cause much change in Future prices.

Therefore the Grangers Causality Test results show that the Future markets are leading the spot

market in impounding future expectations about the future spot price. There is only

unidirectional causality from futures market to Spot market, which helps to conclude that the

Futures markets are more efficient in discovering the future Spot prices.

** Reference Tables No 20 ------- See Annexure

TESTS F- STATISTICS

(SPOT)

P - VALUES AT 1% AT 5% AT 10%

Lag 2 2.21257** 0.10979 1 1 1

Lag 5 0.26713* 0.94586 1 1 1

Lag 12 1.23778* 0.99384 1 1 1

(** indicates acceptance of null hypothesis)



50/73




51/73



ANALYSIS AND INFERENCESThe results of the present study are compared with the studies done by many other researchers.

Ash Narayan and Amit Kumar in their study found out that there exist long run relationship

between nifty spot and nifty futures prices. They found out that the two markets are co-

integrated. The investors can combine information of nifty spot and nifty futures prices to predict

future spot price.

Another study done by Dimitris F Keniurgois (2004) examined the relationship between price

moments of the FTSE/AE 20 futures market. The results showed that two markets are co

integrated. They also found out that there exists bi directional causality between stock index spot

and the futures market series as a focal point of information assimilation and fulfill their price

discovery.

Hector O Tapatra and T Ranball Forters Bery found out that futures market provide a price

discovery function. They also found that there is causal relationship between the two markets.

The present study done by me also shows the results which are similar to past studies. It shows

that there is long run equilibrium relationship between nifty spot index and the nifty index

futures. There exists causal relationship between both the markets. But the past studies have

shown that there in bidirectional causality between the markets, but the present study shown that

there is more of uni directional causality from future to market in feedback mechanism


52/73



CCOONNCCLLUUSSIIOONNSS

The empirical results of the study show that the there exists a long run equilibriumrelationship between Nifty Spot and Nifty futures. This long run relationship between these

two markets is evidenced by the Grangers Co integration Test and Johansen co integration

Test. These tests show positive results towards co integration between Nifty Spot Prices

and Nifty Futures prices.

Further the results of Grangers Causality test indicate that there is lead lag relationship

between Futures market and Spot market. This shows that there is feedback mechanism

between Futures market and Spot markets. But the results significantly show that there is

more of unidirectional causal effect and feedback from Future market to Spot market in

impounding the information in its prices. The occurrence of changes in the futures prices

precede the changes in the Spot prices .This indicates that Futures market is leading the

spot market in discovering the prices for futures Spot prices

So finally we can conclude that the Future market is efficient in discovering the Future

Spot price compared to Spot market


53/73




54/73



BIBLIOGRAPHYREFERNCE BOOKS

Basic Econometrics - Damodar N.Gujarati, (fourth edition)

Options and Futures H C Hull

Market Models -- Indian Institute of Management, Bangalore

WEBSITES www.nseindia.com

www.finance.yahoo.com

www.google.com

www.investorpedia.com

REFERNCE ARTICLES

Efficient Price Discovery in Stock Index Cash and Futures market---By

PASCAL ALPHONSE----2000

Price Discovery in cash and Futures Market---The case of S & P NIFTY and

NIFTY FUTUTRES----Ash Narayan Shah and Anil Kumar A(2006)

Price discovery in the Athens Derivatives Exchange: Evidence for the FTSE/ASE

20 Futures Market---2004

Stochastic interest rates and price discovery in selected commodity markets-----

By Hector. O. Zaptra and T . Randall Fortenbery---1995

Study of common stochastic trend and co integration in the emerging markets

A case study of India, Singapore and Taiwan ---- June 2003
http://www.investorpedia.com/http://www.google.com/http://www.finance.yahoo.com/http://www.nseindia.com/


55/73




56/73



TABLE 01

ADF RESULTS FOR NIFTY SPOT PRICES AT LEVEL, AT NONE, LAG 0

ADF Test Statistic 3.302079 1% Critical Value* -2.5671

5% Critical Value -1.9396


*MacKinnon critical values for rejection of hypothesis of a unit root.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(SPOT)

Method: Least Squares

Date: 06/12/06 Time: 10:46

Sample(adjusted): 1/05/2000 8/08/2005

Included observations: 1459 after adjusting endpoints

Variable Coefficient Std. Error t-Statistic Prob.

SPOT(-1) 0.001129 0.000342 3.302079 0.0010

R-squared 0.003508 Mean dependent var 1.344859

Adjusted R-squared 0.003508 S.D. dependent var 21.42071

S.E. of regression 21.38311 Akaike info criterion 8.963765

Sum squared resid 666652.2 Schwarz criterion 8.967388

Log likelihood -6538.067 Durbin-Watson stat 1.758413



57/73



TABLE 02

ADF RESULTS FOR NIFTY SPOT PRICES AT LEVEL, AT INTERCEPT,

LAG 0








Date: 06/12/06 Time: 10:56

Sample(adjusted): 1/05/2000 8/08/2005



SPOT(-1) 0.003025 0.000993 3.045687 0.0024

C -3.305759 1.626131 -2.032898 0.0422





Log likelihood -6536.000 F-statistic 9.276211

Durbin-Watson stat 1.766748 Prob(F-statistic) 0.002363


58/73



TABLE 03

ADF RESULTS FOR NIFTY SPOT PRICES AT LEVEL, AT TREND

ANDINTERCEPT, LAG 0








Date: 06/12/06 Time: 11:01

Sample(adjusted): 1/05/2000 8/08/2005



SPOT(-1) 0.000163 0.001639 0.099264 0.9209

C -2.419433 1.673461 -1.445766 0.1485

@TREND(1/03/2000) 0.004807 0.002190 2.194787 0.0283








59/73



TABLE 04

ADF RESULTS FOR NIFTY FUTURE PRICES AT LEVEL, AT NONE, LAG 0








Date: 06/12/06 Time: 10:20

Sample(adjusted): 1/05/2000 8/08/2005



SPOT(-1) 0.000813 0.000692 1.174202 0.2405




Sum squared resid 2703200. Schwarz criterion 10.36731

Log likelihood -7559.310 Durbin-Watson stat 2.666096


60/73



TABLE 05

ADF RESULTS FOR NIFTY FUTURE PRICES AT LEVEL, AT

INTERCEPT, LAG 0






Dependent Variable: D(FUTURES)


Date: 06/12/06 Time: 10:42

Sample(adjusted): 1/05/2000 8/08/2005



FUTURES(-1) 0.000594 0.002025 0.293061 0.7695C 0.380141 3.297711 0.115274 0.9082


Adjusted R-squared -0.000627 S.D. dependent var 43.05971






61/73



TABLE 06

ADF RESULTS FOR NIFTY FUTURE PRICES, AT LEVEL, AT

TREND AND INTERCEPT, LAG 0






Dependent Variable: D(FUTURES)


Date: 06/12/06 Time: 10:42

Sample(adjusted): 1/05/2000 8/08/2005



FUTURES(-1) 0.000594 0.002025 0.293061 0.7695

C 0.380141 3.297711 0.115274 0.9082


Adjusted R-squared -0.000627 S.D. dependent var 43.05971






62/73



TABLE 07

ADF RESULTS FOR NIFTY SPOT PRICES AT FIRST DIFFRENCE AT

LEVEL, AT NONE, LAG 0

ADF Test Statistic -33.88683 1% Critical Value* -2.5671





Dependent Variable: D(SPOT1)


Date: 06/12/06 Time: 11:04

Sample(adjusted): 1/05/2000 8/08/2005



SPOT1(-1) -0.881882 0.026024 -33.88683 0.0000

R-squared 0.440589 Mean dependent var 1.77E-05


S.E. of regression 0.013836 Akaike info criterion -5.722349

Sum squared resid 0.279126 Schwarz criterion -5.718726

Log likelihood 4175.454 Durbin-Watson stat 1.971032


63/73



TABLE 08

SHOWING ADF RESULTS FOR NIFTY SPOT PRICES, AT FIRST

DIFFRENCE AT LEVEL, AT INTERCEPT, LAG 0

ADF Test Statistic -33.93323 1% Critical Value* -3.4377





Dependent Variable: D(SPOT1)


Date: 06/12/06 Time: 11:05

Sample(adjusted)

price discovery in cash& futures market-laxmi sn-0483

Documents