Performance Evaluation

FIN 261
Lecture Notes Ch. 18

1
Indirect Investments
Cash
Virtually risk-free money market securities
Passive management
Diversified portfolio with no security mispricing
identification
Active management
Forecasting broad markets and/or identifying mispriced
securities to achieve higher returns
Security selection
Market timing
 Movement between risky portfolio and cash
2
Portfolio Management
Objectives, constraints, and preferences are identified
Expectations about risk and return are needed
Leads to explicit investment policies
Strategies developed and implemented
Market conditions, asset mix, and investor
circumstances are monitored
Portfolio performance is measured and evaluated
Portfolio adjustments are made as necessary

3
Performance Measurement
Allows measurement of the success of portfolio
management
Key part of monitoring strategy and evaluating risks
Important for:
Those who employ a manager
Those who invest personal funds
Could lead to revisions
Find reasons for success or failure
Re-evaluation of risk and return
Costs of restrictions
4
Performance Evaluation
Without knowledge of risks taken, little can be said
about performance
Intelligent decisions require an evaluation of risk and
return
Risk-adjusted performance
Relative performance comparisons
Benchmark portfolio must be legitimate alternative that
reflects objectives
Risk
Total risk
Systematic risk
5
Comparison Groups
Compare with
other funds with
similar risk
characteristics
For example,
comparison
within high-yield
bond portfolios
and within
growth stock
equity funds
6
Risk-adjusted Performance
Must measure normal performance in order to
measure abnormal performance
Single index model (CAPM)
Multi factor model (APT)
Type of risk
Total risk
 When the fund or portfolio will not be mixed
Systematic risk
 When the fund or portfolio will be a small part of larger
portfolios

7
Performance Evaluation Example

8
Sharpe Ratio
 
 SR p  R p  RF /SD p
 Reward-to-variability ratio
Benchmark based on the ex post capital market line
=Average excess return / total risk
Risk premium per unit of TOTAL risk
The higher, the better the performance
Provides a ranking measure for portfolios
For portfolio p in the example
SRp = (13.6 – 4) / 24.1 = 0.398

9
M-square Measure
SR measure not in percentage terms
M-square is the return earned when portfolio's total
risk is either dampened or leveraged to match the
benchmark total risk
Hypothetical riskless borrowing or lending required
 R + (– R )  (σ /σ )
f f M p
-
Or
For portfolio p
M2 = {4+(13.6-4)×18.5/24.1} – 10.4 = 1%

10
M-square of Portfolio P

11
Treynor Index
 TI p  
 R p  RF /β p
Average excess return / market risk
Risk premium per unit of market risk
Example
P’s TI = (13.6 – 4) / 1.25 = 7.68

12
Jensen’s Alpha
 
  p  R p  R f   p RM  R f 
Measures contribution of portfolio manager beyond
return attributable to the market risk
CAPM implies alpha is zero
Often, alpha and beta for a portfolio are estimated
simultaneously from a regression:
Rpt - Rft =ap +bp [RMt - Rft] +ept
Alpha relative to other factor models measures
performance beyond the return predicted by the
model
13
Information Ratio
IRP = Alpha / Residual standard deviation
From single index model R j  R f   j   j ( RM  R f )  e j
 Part of the returns unrelated to the benchmark is  j  e j
 IRP = E[αP + eP] / SD(αP + eP) = αP / SD(eP)
 Average active return / Risk of active return
In the example, IRP = 1.6 / (24.12-1.252×18.52)1/2 = 0.24
Useful when evaluating a portfolio to be mixed with a well
diversified portfolio

14
Which Measure?
Case 1: If the portfolio represents investor’s entire risky
investment fund
 Relevant risk is the total risk
 Use Sharpe ratio or M-squared measure
Case 2: If the portfolio is one of many portfolios combined into
a large investment fund
 Relevant risk is the systematic risk
 Use Treynor index or Jensen’s alpha
Case 3: If the portfolio is to be mixed with a well-diversified
fund
 Portfolio adds active return and risk of active return
 Use information ratio
15
Alpha Capture and Transport
Once found, an alpha can be isolated by hedging the
associated market exposure
Example: 0.5 exposure to XLV desired with the alpha of P

16
Evaluation with Multi-factor Model
•

17
Style Analysis
For actively managed funds, style is an important
determinant of return
E.g., growth / value, small / mid / large cap
Investors seek to compare performance of a fund with
its peer benchmark
Return based style analysis
Use asset class returns as factors
Find which benchmarks fund returns fluctuated with
More than 90% of return variation can be explained by
funds’ allocation to T-bills, stocks, and bonds
18
Style Analysis for Magellan Fund

19
Magellan Fund Cumulative Return
Difference against Style Benchmark

20
Magellan Fund Cumulative Return
Difference against S&P500

21
Morningstar Rating
Holding based rating
Company peer groups established based on
Morningstar style definitions
Risk-adjusted performance ranked and stars assigned
according to table
Percentile Stars
0-10 1
10-32.5 2
32.5-67.5 3
67.5-90 4
90-100 5

22
Market Timing
Adjust asset allocation for movements in market
 Shift between stocks and money market instruments or
bonds
A perfect market timer’s return would look like:

23
Estimation of Market Timing Ability
Add a non-linear term to the estimated characteristic
line such as squared market return term

24
Other Issues
Assumptions underlying measures limit their usefulness
 CAPM / Factor model, and riskless borrowing / lending,
and etc
When the portfolio is being actively managed, basic
stability requirements are usually not met
Sample period issue
Performance attribution (not covered)
 Market timing
 Asset allocation
 Security selection

25