Port.

1 Portfolio Risk and Return: Part II

Learning Module: 2

2. CAPITAL MARKET THEORY: RISK-FREE AND RISKY ASSETS

Portfolio of Risk-Free and Risky Assets & &+(

Stanley Notes™ 2 0 2 5
&
𝜎! = ., 𝑤%# 𝜎)# +,, 𝑤% 𝑤) 𝜎% 𝜎) 𝜌
)'%*(
%'( %'(
According to Risk-return trade-off, higher returns can
only be achieved by taking on greater risk. The
where,
investment opportunity set can be expanded towards
the northwest, indicating an improved portfolio by E (Rp) = expected return on the portfolio
adding assets with correlation <+1 with the existing asset wi = fractional weight in asset i
class. Ri = expected return of asset i

For example, combination of risk-free asset and a risky The covariance between asset is and j, Cov (i,j) is
asset (known as Capital allocation line) provides better expressed as follows:
risk-return trade-off because correlation between risk-
free asset and risky asset is zero. Covariance between asset i &j = Cov (i, j) = ρij×σi×σj
Correlation coefficient (ρ) is expressed as:
Combining a Risk-Free Asset with a Portfolio of
Risky Assets COVij
r ij =
Expected return of a portfolio (with two assets) is s isj
estimated as follows:
Since the correlation of an asset with itself is 1,
E(rA) = wAE(rA)+wBE(rB)
where, Cov (i, j) = ρii×σi×σi = σ2i
A = first asset
B = second asset Assume three portfolios of risky assets, A, B, and C.
w = weights (respectively)
E(r) = expected return of assets • We can create three CALs by combining each
portfolio with the risk-free asset i.e. CAL (A), CAL
NOTE: (B), CAL (C).
• Among portfolios A, B and C, the portfolio C is the
It is assumed that portfolio contains all available risky most superior and is preferred by investors
assets. When any asset is not included in the portfolio, its because it has a greater expected return for any
weight will be = 0. given level of risk.
• The optimal allocation along the capital
Risk (S.D.) of a portfolio (with two assets) is given by: allocation line depends on the risk-aversion of
investors.
𝜎! = #𝑤"# 𝜎"# + 𝑤$# 𝜎$# + 2𝑤" 𝑤$ 𝜎" 𝜎$ 𝜌"$ o A risk-averse investor may prefer to invest a large
proportion of his/her assets in the risk-free asset
where, and earns low returns (due to lower risk
assumed).
A = first asset o A risk-seeking investor may prefer to invest a
B = second asset large portion of his/her assets in the risky asset
w = weights (respectively) and earns higher return (due to higher risk
σ = standard deviation of assets assumed).
r = correlation coefficient of the two assets

Expected return and risk for a portfolio with N number of

assets:

Portfolio return can be stated as:

&

E(𝑟! ) = , 𝑤% 𝐸(𝑟% )
%'(

Portfolio risk can be stated as:

–––––––––––––––––––––––––––––––––––––– Copyright © FinQuiz.com. All rights reserved. ––––––––––––––––––––––––––––––––––––––

 Port.1 Portfolio Risk and Return: Part II
Learning Module: 2

Does a Unique Optimal Risky Portfolio Exist?

When investors have homogeneous expectations (i.e.

same economic expectations and same expectations of
prices, CFs etc.), then only one optimal risky portfolio
exists.

In case of different expectations, investors would have

different estimates of returns, risk and required rates of
return resulting in different optimal risky portfolios.
Investors can use market prices and the market portfolio
as a base case or benchmark to evaluate performance
of other portfolios.

3. CAPITAL MARKET THEORY: THE CAPITAL MARKET LINE

Passive and Active Portfolios The Capital Market Line (CML)

The CML is a special case of the CAL. In CML,

Passive Portfolios: Passive portfolios are based on the
assumption of unbiased market prices. They replicate
and track market indices e.g. S&P500 index, Nikkei 300, • The risky portfolio is represented by a market
and CAC 40. They are also known as index funds. portfolio.
o The S&P500 index can be used as a proxy of the
market portfolio. Thus,
• Passive portfolios have low costs relative to active
The return on risky portfolio = Expected market
portfolios.
return i.e. E (Rm).
• The risk-free asset is a debt security with no default
Active Portfolios: Active portfolios are based on investor’s risk, no inflation risk, no liquidity risk, no interest rate
forecasts regarding cash flows, growth rates and risk, and no other risk.
discount rates. In an actively managed portfolio, o Return on U.S. Treasury bills can be used as a
proxy of the risk-free return Rf.
• Undervalued (overvalued) assets have a positive
(negative) weight i.e. they are over-weighted The equation of CML is:
(under-weighted) relative to the market weight in
the benchmark index. 𝐸(𝑅. ) − 𝑅-
• Most open-end mutual funds and hedge funds 𝐸/𝑅, 1 = 𝑅- + 2 4 × 𝜎,
𝜎.
are actively managed.
• Intercept = risk-free rate
• Slope (known as the market price of risk) = [E(Rm) –
What is the Market? Rf)] / σm.
o èThe slope of CML is positive because the
Theoretically, the market includes all risky assets e.g. market’s risky return > risk-free return.
stocks, bonds, real estate, human capital etc. However, o è Both S.D. and expected return increase with
some assets are non-tradable and/or non-investable. the increase in the amount invested in the
market portfolio.

• Example of Non-tradable assets: Human capital.

• Example of Non-investable assets: Class A shares
on the Shanghai stock exchange are available to
domestic investors i.e. foreign investors cannot
invest in Class A shares.

The S&P500 Index can be used by investors as a market’s

proxy. S&P500 Index is a market-capitalization weighted
index and contains 500 of the largest stocks (domiciled
in the U.S.).
 Port.1 Portfolio Risk and Return: Part II
Learning Module: 2

correlated); whereas a combination of the risk-

free asset and a risky portfolio is a straight line.
• Only well-diversified portfolios lie on the CML.

Expected return of a portfolio consisting of risk-free asset

and risky asset:

E(Rp) = w1Rf+(1-w1) E(Rm)

S.D. of a portfolio consisting of risk-free asset and risky

asset:
𝝈𝒑 = #𝝎𝟐𝟏 𝝈𝟐𝒇 + (𝟏 − 𝝎𝟏 )𝟐 𝝈𝟐𝒎 + 𝟐𝝎𝟏 (𝟏 − 𝝎𝟏 )𝑪𝒐𝒗/𝑹𝒇 , 𝑹𝒎 1

where,
As shown in the fig above,

• S.D. or total risk is plotted on the x-axis. w1 = proportion of wealth invested in the risk-
• Portfolio’s Expected return is plotted on y-axis. free asset
• The Y-intercept is the risk-free return (Rf). 1- w1 = proportion of wealth invested in the
• The market portfolio is represented by the point on market portfolio (risky asset)
the Markowitz efficient frontier. It represents the σf = risk of risk-free asset
optimal risky portfolio. σm = risk of the market portfolio
o All points that lie under (or below) CML are not σp = risk of portfolio
efficient portfolios because they generate the Cov (Rf, Rm) = covariance between the risk-free asset
same level of return with a higher level of risk or and the market portfolio
lower level of return with the same amount of
risk. Since, the S.D. of the risk-free asset is 0, the portfolio
o All points that lie above CML are not attainable. return and S.D. can be simplified as follows:
o All points that lie on the CML represent efficient
portfolios. E(Rp) = w1Rf+(1-w1) E(Rm)
o The optimal portfolio is the point of tangency
between CML and Markowitz efficient. 𝝈𝒑 = (𝟏 − 𝝎𝟏 )𝝈𝒎

Important to note:

Practice: Example 1 from the

• The combination of two risky assets is usually not a CFA Institute’s Curriculum.
straight line (except when the assets are perfectly

4. CAPITAL MARKET THEORY: CML – LEVERAGED PORTFOLIOS

Leveraged Portfolios • At point Rf, all of the wealth is invested into risk-free
securities. In other words, investor lends 100% of
his/her total wealth at risk-free rate.
• At point M, the investor invests 100% of his/her total
wealth in the market portfolio.

Lending Portfolios: All portfolios that lie between points Rf

and M are referred to as ‘lending’ portfolios.

Borrowing Portfolio: Assuming that lending and

borrowing rates are same, an investor can borrow at the
risk-free rate and invest that amount in risky asset (M) i.e.
weight of market portfolio >1.00 and weight of risk-free
asset < 0 All portfolios that lie on a straight line to the
right of point M (i.e. extended portion of the straight line
 Port.1 Portfolio Risk and Return: Part II
Learning Module: 2

joining Rf and M) are referred to as “borrowing”

portfolios.

• Negative weight in risk-free asset implies that an

investor takes a “leveraged” position in the risky
portfolio.
• The optimal allocation along the CML depends on
the risk-aversion of investors.

Practice: Example 2 from the

CFA Institute’s Curriculum.

As shown in the fig above,

Leveraged Portfolios with Different Lending and
Borrowing Rates • When lending rate = borrowing rate, slope of
CML= [E(Rm) – Rf)] / σm between points Rf and M.
Assume: • When borrowing rate > lending rate, slope of CML
will be smaller at points to the right of M.
• Lending rate = risk-free rate i.e. Rf
• Borrowing rate = Rb è it is > risk-free rate. CML equation when the investment in the risk-free asset
≥ 0.
When lending rate is not the same as borrowing rate, the
𝐸(𝑅. ) − 𝑅-
CML will no longer be a single straight line; rather CML 𝑤( ≥ 0: 𝐸/𝑅, 1 = 𝑅- + 2 4 × 𝜎,
will be kinked at point M. 𝜎.

CML equation when the investment in the risk-free asset

• All passive portfolios will lie on the kinked CML.
< 0.
𝐸(𝑅. ) − 𝑅4
𝑤( < 0: 𝐸/𝑅, 1 = 𝑅4 + 2 4 × 𝜎,
𝜎.

Practice: Example 3 from the

CFA Institute’s Curriculum.

5. SYSTEMATIC RISK AND NONSYSTEMATIC RISK

Total Variance = Systematic Variance + Nonsystematic

Systematic Risk and Nonsystematic Risk
Variance

Total Risk = Systematic risk + Nonsystematic risk

Systematic risk/non-diversifiable/market risk: It is the risk

that affects the entire market or economy and cannot
be reduced through diversification. Examples include risk
associated with interest rates, inflation, economic cycles,
political uncertainty and widespread natural disasters.

Nonsystematic risk or diversifiable risk: It is the risk that

affects a single company or industry. It is also known as
company-specific, industry-specific or idiosyncratic risk
e.g. failure of a drug trial, major oil discoveries or an
airliner crash. Nonsystematic risk can be reduced or
avoided through diversification i.e. investing in different
asset classes, securities.
 Port.1 Portfolio Risk and Return: Part II
Learning Module: 2

current price is expected to have a higher (lower)

• The figure shows that portfolio’s risk decreases with future return.
the increase in the number of stocks in a portfolio.
• The maximum reduction in portfolio’s risk can be In an efficient market, investors can earn incremental
achieved using 30 stocks; beyond that point, the return for assuming systematic risk only, implying that the
S.D. remains constant and the only remaining risk only risk that is priced is the systematic risk. Therefore, risk-
is systematic risk. averse investors should hold only well-diversified
portfolios.
Pricing of Risk
Pricing or valuing an asset involves estimating the
expected rate of return of an asset. Practice: Example 4 from the
CFA Institute’s Curriculum.
• An asset with a known terminal value (e.g. face
value of a bond): An asset with a lower (higher)

6. RETURN GENERATING MODELS

Multi-factor model is expressed as follows:

Return Generating Models
𝒌 𝒌

𝑬(𝑹𝒊 ) − 𝑹𝒇 = , 𝛃𝒊𝒋 𝑬(𝑭𝒋 ) = 𝛃𝒊𝐥 F𝑬(𝑹𝒎 ) − 𝑹𝒇 G + , 𝛃𝒊𝒋 𝑬(𝑭𝒋 )

A return-generating model is a model that is used to 𝒋'𝟏 𝒋'𝟐
estimate the expected return of a security given certain where,
parameters (e.g. with given level of systematic risk). For
k = number of factors è these represent risk
example,
factors that would provide investors the
required return or premium for assuming that
Parameters estimates are obtained by estimating past
risk.
relationship between various factors and returns è then
βij = factor weights or factor loadings associated
the estimated parameters are used to estimate
with each factor
expected returns.
E (Ri – Rf) = excess return or return over the risk-free rate
E (Rm) = expected market return
Limitation of the model: The accuracy of estimates of
expected return depends on the quality of input
Three-Factor and Four-Factor Models:
estimates and the accuracy of the model.
A three and four factor return-generating models
A multi-factor model is the most general form of a return- include following factors:
generating model. A multi-factor model involves various
kinds of factors i.e. macroeconomic, fundamental and • Relative size of the company
statistical factors. • Relative book-to-market value of the company
• Beta of the company.
• Macroeconomic factors include economic • Momentum i.e. relative past stock returns.
growth, interest rate, inflation rate, productivity,
employment, and consumer confidence etc. The Single-Index Model: It is the simplest form of a return-
• Fundamental factors include i.e. earnings, generating model. As the name suggests, it is based on
earnings growth, CF generation, investment in only one factor (i.e. the market factor).
research, advertising and number of patents etc.
• Under Statistical models, historical and cross- E (Ri) – Rf = βi × [E(Rm) –Rf)]
sectional return data are analyzed to determine
factors upon which the variance or covariance of Or
observed returns depend.
𝜎% β% 𝜎.
𝐸(𝑅% ) − 𝑅- = H I × F𝐸(𝑅. ) − 𝑅- G = H I × [𝐸(𝑅. ) − 𝑅- ]
Limitation of statistical models: These models may have 𝜎. 𝜎.
data mining problem e.g. spurious correlation may exist
between two variables without any economic meaning. where,
σi / σm = (Total security risk / Total market risk) = Factor
loading or factor weight
 Port.1 Portfolio Risk and Return: Part II
Learning Module: 2

Decomposition of Total Risk for a Single-Index Return-Generating Models: The Market Model
Model

Market model is the same as the single-index model. The

The single index model can be expressed in the following market model is expressed as:
using realized returns rather than expected returns. Ri = αi + βi Rm +ei

Ri – Rf = βi(Rm – Rf) + ei
where,
Where, ei = error term that reflects non-market changes αi = Intercept
We know, βi= slope coefficient

Total variance = systematic variance +Nonsystematic NOTE:

variance
Intercept and slope coefficients are estimated by using
σ2i = β2i σ2m + σ2e historical security and market returns.

Total S.D. is given by: Example:

A company’s historical daily returns are regressed on
si = b is 2 2
m +s 2
e
S&P500 daily returns to estimate intercept and slope i.e.

where, • αi = 0.0001
• βi = 0.90
βi × σm = systematic risk
σe = Non-systematic risk
Thus, regression equation can be stated as follows.
A well-diversified portfolio (i.e. the market portfolio) has Company’s expected daily returns = 0.0001 + 0.90 × Rm
no nonsystematic risk. Thus,
Assume, Rm = 1% and stock rises by 2%, then the
Total risk of for a well-diversified portfolio = Systematic risk abnormal return or company specific return is computed
=βi×σm as follows.

Company specific return (ei) = 0.02 – [0.0001 + (0.90 ×

0.01)] = 1.09%

7. CALCULATION AND INTERPRETATION OF BETA

Beta is used to measure the sensitivity of an asset’s return Example:

to the market as a whole. Beta reflects the systematic risk
of an asset. • Correlation between an asset and the market =
0.70
• When an asset has positive beta (β> 0), it indicates • S.D. of Asset = 0.25
that the return of an asset moves in the same • S.D. of market = 0.15
direction of the market.
• When an asset has negative beta (β< 0), it Asset’ beta = (0.70 × 0.25) / 0.15 = 1.17
indicates that the return of an asset moves in the
opposite direction of the market. when
• When an asset has beta = 0, it indicates that the
movement in asset’s return is unrelated with • Asset’s covariance with the market = 0.026250
movements in the market. • Market variance = 0.02250

𝐶𝑜𝑣(𝑅% , 𝑅. ) 𝜌%,. 𝜎% 𝜎. 𝜌%,. 𝜎% Asset’s Beta = 0.026250 / 0.02250 = 1.17

𝛽% = # = # =
𝜎. 𝜎. 𝜎.

The beta of the market = 1 i.e.

Practice: Example 5 from the
𝜌%,. 𝜎% 𝜌.,: 𝜎. CFA Institute’s Curriculum.
𝛽% = = =1
𝜎. 𝜎.
 Port.1 Portfolio Risk and Return: Part II
Learning Module: 2

Estimation of Beta Issues involved in beta estimation:

Beta and Expected Return
• Beta that is estimated using shorter period data
Beta can also be estimated directly by using the market (e.g. 12 months) more appropriately reflect asset’s
model. current level of systematic risk relative to longer
Ri = αi + βiRm + ei period beta estimates (e.g. 3-5 years).
• However, longer period beta estimates are more
• The two parameters of the regression (αi and βi) accurate relative to shorter period beta estimates.
are estimated by using historical security returns
(Ri) and historical market returns (Rm) as inputs into
the regression model.
Practice: Example 6 from the
CFA Institute’s Curriculum.

CAPITAL ASSET PRICING MODEL: ASSUMPTIONS AND THE

8.
SECURITY MARKET LINE

CAPM is an equilibrium model. CAPM equation is written Implication:

as:
If this assumption is not satisfied, it has no significant
effect on the general conclusions of the CAPM.
E (Ri) = Rf + βi × [E (Rm) – Rf)]
However, costs of short selling or restrictions on short-
According to CAPM, systematic risk, measured by asset’s selling can result in an upward bias in asset prices and
beta, is the primary determinant of expected return for a may affect conclusions of the CAPM.
security i.e. when two assets have equal betas, their
expected returns will be the same regardless of the 3) Investors plan for the same single holding period: The
nature of those assets. CAPM is based on the assumption of single holding
period.
• The higher (lower) the beta of an asset, the higher
(lower) its expected return will be. Implication:
• When beta of an asset > (<) 1, è asset’s However, the CAPM can easily be applied for multi
expected return > (<) market return. holding periods.
• When asset’s beta < 0 (i.e. negative) è asset’s
required return < risk-free rate. 4) Investors have homogeneous expectations about
asset returns:
NOTE:
Implication:
Insurance is an asset with negative beta.
CAPM can be applied if investors have heterogeneous
expectations unless different expectations do not
Assumptions of the CAPM
generate significantly different optimal risky portfolios.

1) Investors are risk-averse, utility-maximizing, rational 5) All investments are infinitely divisible: An investor is
individuals: able to invest as little or as much as he/she wishes in
an asset.
• Utility maximization implies that investors prefer
higher returns, not lower returns. Implication:
• Rational investors imply that investors However, violation of this assumption does not have any
appropriately evaluate and analyze all available significant impact on the conclusions of the model.
information to make rational decisions.
6) Investors are price takers: There are many investors
Implication: and no investor is able to influence prices i.e. investors
Investors’ irrational behavior does not affect CAPM’s are price takers.
results unless prices are significantly affected.
The Security Market Line
2) Markets are frictionless, including no transaction costs
and no taxes: Frictionless markets refer to markets,
The security market line (SML) is a graphical
which have no transaction costs, taxes, or any costs
representation of the CAPM.
or restrictions on short-selling.
 Port.1 Portfolio Risk and Return: Part II
Learning Module: 2

If a point representing the estimated return of an asset is

plotted above the SML, it indicates that the asset is Reason:
undervalued i.e. has a low level of risk relative to the
amount of expected return. This asset should be • The CAL and CML use the total risk of the asset
purchased. rather than its systematic risk. Therefore, CAL and
CML can only be applied to those assets whose
If the point representing the estimated return of an asset total risk is equal to systematic risk i.e. efficient
is plotted below the SML, it indicates that the asset is portfolios.
overvalued i.e. has a higher level of risk relative to the • In contrast, SML is based on asset’s systematic risk,
amount of expected return. not its total risk.

NOTE:
Practice: Example 7 from the
If short selling is allowed, then such an asset should be
CFA Institute’s Curriculum.
short sold.

All properly valued securities and portfolios (both

diversified and non-diversified) will be plotted along the 8.2.1) Portfolio Return and Portfolio Beta
SML in equilibrium. They will have beta = 1 and expected
Portfolio expected return can be estimated as follows:
return equal to the expected return on the market.
𝐸(𝑅, ) = 𝑅- + 𝛽, F𝐸(𝑅. ) − 𝑅- G
Or
𝐸(𝑅( ) = 𝑅- + 𝛽( F𝐸(𝑅. ) − 𝑅- G
𝐸(𝑅# ) = 𝑅- + 𝛽# F𝐸(𝑅. ) − 𝑅- G
𝐸/𝑅, 1 = 𝑤( 𝐸(𝑅( ) + 𝑤# 𝐸(𝑅# )
= 𝑤( 𝑅- + 𝑤( 𝛽( F𝐸(𝑅. ) − 𝑅- G + 𝑤# 𝑅-
+ 𝑤# 𝛽# F𝐸(𝑅. ) − 𝑅- G
= 𝑅- + (𝑤( 𝛽( + 𝑤# 𝛽# )F𝐸(𝑅. ) − 𝑅- G

The portfolio beta is a weighted sum of the betas of the

component securities i.e.
: :

𝛽, = , 𝑤% 𝛽% ; , 𝑤% = 1
As shown in the graph above, %'( %'(

• Beta is plotted on the x-axis. Important to Note:

• Expected return is plotted on the y-axis. CAPM and SML are used to determine the expected
• The Y-intercept of SML is the risk-free rate of return return in the market given a certain level of risk. It will not
(Rf). be necessarily the same as the actual or realized return.
• The slope of SML is the market risk premium (Rm –
Rf).

Practice: Example 8 from the

SML versus CAL and CML:
CFA Institute’s Curriculum.

• The CAL and the CML can only be applied to

efficient portfolios.
• The SML can be applied to any security (both
efficient and inefficient).

9. CAPITAL ASSET PRICING MODEL: APPLICATIONS

Estimate of Expected Return • Expected return obtained from CAPM can be

used for valuing assets i.e. stocks, bonds, real
estate etc.
CAPM can be used to estimate the expected return for • The required rate of return obtained from CAPM
capital budgeting purposes: For example,
can be used to determine the economic
feasibility of capital projects.
 Port.1 Portfolio Risk and Return: Part II
Learning Module: 2

• The required rate of return obtained from CAPM

can be used to calculate the cost of capital or to Practice: Example 9 from the
set fair insurance premiums. CFA Institute’s Curriculum.

10. BEYOND CAPITAL: LIMITATIONS AND EXTENSIONS OF CAPM

5) CAPM assumes homogeneous investor expectations:

The implication of homogeneous expectations under
Limitations of the CAPM
CAPM is that there is a single(same) optimal risky
portfolio (market) and a single(same) security market
CAPM is based on constraining and unrealistic line for all investors.
assumptions. The CAPM is subject to theoretical and
practical limitations.
Extensions to the CAPM
i) Theoretical Limitations of the CAPM
i) Theoretical Models
1) CAPM is a Single-factor model: Under CAPM,
systematic risk (measured by asset’s beta) is the An arbitrage pricing theory (APT) is a form of theoretical
primary determinant of expected return for a security. model. Like the CAPM, APT exhibits a linear relationship
This implies that only systematic risk or beta risk is between expected return and risk. It is expressed as:
priced. It makes CAPM a restrictive and inflexible
model. 𝐸/𝑅, 1 = 𝑅; + 𝜆< 𝛽,,= + ⋯ + 𝜆> 𝛽,,>

2) CAPM is a Single-period model: The CAPM does not where,

consider multi-period implications or investment
objectives of future periods. E (Rp) = expected return of portfolio p
Rf = risk-free rate
ii) Practical Limitations of the CAPM λj = risk premium (i.e. expected return in excess of
the risk-free rate for factor j)
βpj = sensitivity of the portfolio to factor j
1) Market portfolio: Under CAPM, the true market K = number of risk factors
portfolio includes all financial and nonfinancial assets.
However, some assets (e.g. human capital and assets Under APT, the risk factors and betas are estimated on
in closed economies) are not investable. In addition, the basis of no-arbitrage condition in asset markets.
the true market portfolio of CAPM is unobservable.
• Unlike the CAPM, APT involves various risk factors.
2) Proxy for a market portfolio: As the true market • Except the risk-free rate, the risk factors may vary
portfolio is not observable, proxies are used to from one asset to another.
represent market portfolio, which vary among
analysts, investors, countries resulting in different return
estimates for the same asset, which is not allowed in Advantages of APT: Theoretically, APT is elegant, flexible,
the CAPM. and superior to the CAPM.

3) Estimation of beta risk: Limitations of APT:

• In order to have an accurate estimate of beta risk, • APT is not commonly used.
it is computed using a long history of returns (3-5 • APT does not specify risk factors to be used as
years). However, longer period data does not inputs into the model. Therefore, practically,
appropriately reflect asset’s current level of CAPM is superior to APT.
systematic risk.
• Using different periods for estimation results in ii) Practical Models
different estimates of beta and as a result different
estimates of expected return for the same asset.
A four-factor model is an example of practical model. It
better predicts asset returns relative to the CAPM. Four
4) The CAPM is a poor predictor of returns: The CAPM is factors include:
an ex-ante model; but it is based on ex-post data. In
addition, asset returns cannot be solely determined 1) Firm Size i.e. smaller companies outperform larger
by its systematic risk. companies.
 Port.1 Portfolio Risk and Return: Part II
Learning Module: 2

2) Firm book value-to-market value ratio i.e. value • HML = high minus low = difference in returns
companies outperform growth companies. between high-book-to-market- stocks and low-
3) Excess return on the market portfolio (Rm – Rf) book-to-market stocks è it reflects value versus
4) Price momentum i.e. past winners outperform growth stocks.
past losers. • UMD = difference in returns of the prior year’s
winners and losers è It reflects momentum.
The model is expressed as follows: Historically, it has been analyzed that stock return is
unrelated to the market (i.e. βi, MKT is not significantly
𝑬(𝑹𝒊𝒕 ) = 𝛼% + 𝛽%,@AB 𝑀𝐾𝑇C + 𝛽%,D@$ 𝑆𝑀𝐵C + 𝛽%,E@F 𝐻𝑀𝐿C different from 0).
+ 𝛽%,G@H 𝑈𝑀𝐷C
Limitations of the model:
where,
• No strong economic arguments exist for the three
• E (Ri) = return on an asset in excess of the 1-month additional risk factors used by the model.
T-bill return • The model cannot necessarily be applied to other
• MKT = excess return on the market portfolio assets or assets in other countries.
• SMB = small minus big = difference in returns • The model does not necessarily continue to work
between small-capitalization stocks and large- well in the future.
capitalization stocks è It reflects size.

11. PORTFOLIO PERFORMANCE APPRAISAL MEASURES

Performance Evaluation is termed as measuring, to be compared with the Sharpe ratio of another
attributing and evaluating the investment results. portfolio.
• It does not provide any information regarding the
Compare actual return of a portfolio or portfolio economic significance of differences in
manager with the CAPM return for performance performance.
appraisal purposes: • It cannot be used to evaluate portfolio’s
performance relative to the passive market
Portfolio performance can be evaluated using following portfolio.
ratios.

2. Treynor Ratio
1. Sharpe Ratio

𝑅, − 𝑅-
Sharpe Ratio (also known as reward-to-variability ratio): It 𝑇𝑟𝑒𝑦𝑛𝑜𝑟 𝑟𝑎𝑡𝑖𝑜 =
𝛽,
is the slope of CAL.
𝑅, − 𝑅-
𝑆ℎ𝑎𝑟𝑝𝑒 𝑟𝑎𝑡𝑖𝑜 = • Unlike Sharpe ratio, it is based on systematic
𝜎,
(beta) risk.
• In order to have meaningful results, both the
• The greater (lower) the ratio, the better (worse)
numerator and denominator must be positive.
the portfolio’s performance.
• It is based on the total risk of an asset. Therefore, it
is appropriate to use when the portfolio represents Limitations:
an investor’s total portfolio.
• This ratio is applicable only when all comparison • It cannot be used for negative beta assets.
portfolios have positive numerator value. When • It does not provide any information regarding the
the numerator is negative, the ratio results in economic significance of differences in
incorrect rankings. performance.
• It cannot be used to evaluate portfolio’s
Advantages: It is easy to compute. performance relative to the passive market
portfolio.
Limitations:

• It uses total risk as a measure of risk; but only 3. M-squared (M2)

systematic risk is priced.
• It is not informative because in order to rank M2 represents the mean incremental return over a
portfolios, the Sharpe ratio of one portfolio needs market index of a hypothetical portfolio, which is
 Port.1 Portfolio Risk and Return: Part II
Learning Module: 2

created by combining the account with borrowing or Alpha = Actual return on the portfolio during evaluation
lending (leveraging or de-leveraging, respectively*) at period – Expected return on the portfolio given
the risk-free rate so that its standard deviation is identical its systematic risk (measured by its beta)*
to that of the market index.
𝛼, = 𝑅, − F𝑅- + 𝛽, /𝑅. − 𝑅- 1G
𝜎.
M# = /R ! − R- 1 − /𝑅. − 𝑅- 1
𝜎, NOTE:
Interpretation:
By definition, alpha of the market = 0
• When M2 is positive (negative), it indicates that
All returns in the equation are realized, actual returns.
the portfolio outperformed (underperformed) the
market on a risk-adjusted basis.
Interpretation: When αp is positive (negative), it indicates
• When M2 = 0, it indicates portfolio’s return =
that portfolio has outperformed (underperformed) the
market return.
market.

NOTE:
• Like Treynor ratio, it is based on systematic risk.
M2 is stated in percentage terms. • It also reflects the maximum amount that an
investor should be willing to pay the manager to
• * When risk of a portfolio is lower (greater) than the manage his/her money.
market index, leverage (de-leverage) is used and
the hypothetical portfolio is compared to the
market.
• Like Sharpe ratio, it is based on total risk.
• Both Sharpe ratio and M2 provide the similar
rankings.

Example: Suppose,

Portfolio A’s alpha = 5%

Portfolio B’s alpha = 2%.

It indicates that Portfolio A outperformed Portfolio B by

3% and Portfolio A outperformed the market by 5%.
4. Jansen’s Alpha

The Jensen’s alpha or ex post alpha measures the excess Practice: Example 10 from the
of the portfolio’s return over that predicted by the CAPM CFA Institute’s Curriculum.
during the evaluation period i.e.

12. APPLICATIONS OF THE CAPM IN PORTFOLIO CONSTRUCTION

estimated by regressing the excess security return (Ri – Rf)

on the excess market return (Rm – Rf).
Security Characteristic Line
The equation of SCL is:
Security characteristic line (SCL) exhibits the linear
relationship between the return on individual securities 𝑅% − 𝑅- = 𝛼% + 𝛽% /𝑅. − 𝑅- 1
and the overall market at every point in time. It can be
• Intercept = Jensen’s alpha
 Port.1 Portfolio Risk and Return: Part II
Learning Module: 2

• Slope = beta In addition, securities already included in the market

portfolio may be undervalued or overvalued based on
investor expectations.

• When αi is positive, security is undervalued and

investors should increase investment in that
security.
• When αi is negative, security is overvalued; thus, it
should be excluded from the market portfolio (if it
is possible to exclude individual securities).
• The higher (lower) the value of security’s alpha,
the greater (lower) the security’s weight should
be.
• The greater (lower) the nonsystematic risk of a
security, the lower (greater) the security’s weight
should be.
Security Selection
Important to Note:
Analyze alternate return estimates and the CAPM returns
for security selection purposes: • Even a correctly priced security may have non-
zero αi.
The CAPM-calculated price reflects the current market • The weight of the security that is not a part of
price of an asset because it is based on the market portfolio (i.e. non-market security) should
expectations of all other investors in the market. be proportional to:
𝐀𝐥𝐩𝐡𝐚 𝐨𝐟 𝐒𝐞𝐜𝐮𝐫𝐢𝐭𝐲 𝐢
𝐍𝐨𝐧𝐬𝐲𝐬𝐭𝐞𝐦𝐚𝐭𝐢𝐜 𝐯𝐚𝐫𝐢𝐚𝐧𝐜𝐞 𝐨𝐟 𝐒𝐞𝐜𝐮𝐫𝐢𝐭𝐲 𝐢
= αi / σ2i
• When the investor-estimated current price is
higher (lower) than the market price, the asset is • The total weight of nonmarket securities in the
considered undervalued (overvalued). portfolio should be proportional to:
• A positive (negative) Jensen’s alpha indicates ∑&
%'( 𝜔% 𝛼%
that a security is expected to outperform
∑%'( 𝜔%# 𝜎%#
&
(underperform) the market on risk-adjusted basis.
• The weight in the market portfolio is a function of:
Important to note: 𝐸(𝑅. )
#
Portfolios should be evaluated using the appropriate 𝜎.
benchmarks e.g. it is not appropriate to evaluate a real
estate fund against the S&P 500 Index. In addition, the • Information ratio: It is used to measure the
risk and return of the market portfolio and the portfolios security’s abnormal return per unit of risk.
being evaluated should be estimated carefully. Information ratio = αi / σi = (Alpha of security i /
Nonsystematic risk of security i)
• The larger (smaller) the information ratio, the
Implications of the CAPM for Portfolio Constructing
better (worse) the security’s performance.

The S&P500 index can be used as a base portfolio. It

represents the market portfolio for the CAPM. Any
security that is not included in the S&P500 index can be Practice: Example 11 from the
included in the portfolio on the basis of its alpha, which is CFA Institute’s Curriculum.
calculated using the CAPM i.e.

• When αi is positive, security should be added to

Practice: CFA Institute’s end of
the portfolio.
Chapter Questions and Questions
• When αi is negative, security should be short sold.
from FinQuiz Question Bank.