getting started using ibbotson attribution software: an unofficial tutorial mgt 544 stanley...
TRANSCRIPT
Getting started using Ibbotson Attribution software:
An unofficial tutorial
MGT 544
Stanley Martinez, TA
03.08.01
Agenda:
1. Why this is important
2. Encorr Attribution
a. Selecting a series/ benchmark
b. Rolling style analysis
c. Attribution
d Regression
e. Creating a portfolio of managers
1. Why this is important
• Alpha may not be a suitable criteria in a diversified portfolio
• Manager characteristics may not be stable over time
• Managers with similar strategies may exhibit varying correlations, risks and returns
• Manager results may be attributable to alternate factors
Ibbotson Attribution allows analysts to answer these questions
Methodology: “Returns based style analysis”, a constrained form of regression
2. EnCorr Attribution
Main menu (at right)
a. Select portfolios from “Manager/Portfolios” folder
In this example, select:
Raw Data Piper Universe Commingled Value/ Large Cap Value Key Asset Management EB/
Select benchmarks from “Benchmarks” at folder
In this example select all:
Raw Data Ibbotson database Sample Domestic Equity Database:
Your result will include
Large Growth, Large Value, Small Growth, Small Value, Midcap, Microcap & Cash
2. Encorr Attribution
b. Select Style
In this example, the style of the selected managed portfolios should default to 3/88 to 12/98
One can also adjust the regression to “maximum accuracy” if desired
2. Encorr Attribution
Results:
1. Average Style:
Results at right show results of a regression with cumulative weights forced to 100%
2. Rolling style
Results at right show how the weights have changed over time
Large Growth
Large Value
Mid Cap Growth
Mid Cap Value
Small Growth
Small Value
Micro Cap Cash R Squared
Key Asset Management Inc EB
Value Equity
30.02 59.51 0.00 0.00 0.00 0.00 3.59 6.88 96.93
Time
WeightsKey Asset Management Inc EB Value Equity - Estimated Weights
Rolling Style Distribution
0.0%
100.0%
5.0%
10.0%
15.0%
20.0%
25.0%
30.0%
35.0%
40.0%
45.0%
50.0%
55.0%
60.0%
65.0%
70.0%
75.0%
80.0%
85.0%
90.0%
95.0%
Mar1993
Dec1998
Jun1993
Dec1993
Jun1994
Dec1994
Jun1995
Dec1995
Jun1996
Dec1996
Jun1997
Dec1997
Jun1998
Large Growth Large Value Mid Cap Growth Mid Cap Value Small Growth Small Value Micro CapCash
2. Encorr Attribution
c. Select Attribution
Decomposes manager returns into
• Policy: returns due to the investor’s asset allocation decision
• Timing: returns due to investing in the right asset class at the right time
• Selection: returns due to the managers ability to select individual securities within an asset class
In this example: Key’s Attribution Policy: 18.33
Timing: -0.79
Selection 1.39
2. Encorr Attribution
d. Select Regression
Determine how statistically significant the manager’s returns are relative to their benchmark
In this example, Key’s regression results are summarized below
Alpha Alpha T Statistic
Beta Beta T Statistic
R Squared
Number of Observations
Key Asset Management Inc EB Value Equity
0.0025 1.2998 0.9984 36.4293 0.9693 44
2. Encorr Attribution
e. Market Capture
How well is a manager performing when her benchmark is moving up or down?
“Thank you for completing this tutorial”