Portfolio Choice

CAPM — Passive strategies, Index Funds, ETFs

Mehran Ebrahimian
Investment Management (BE452)
Spring 2025
Learning Goals

Basics of portfolio choice:

▶ Two-fund separation, challenges
▶ CAPM, economic foundations
▶ Passive investment strategies
▶ Index Funds, ETFs
Review — Modern Portfolio Theory

Markowitz’s model
▶ “Mean-variance investor”
▶ Frictionless trading
▶ Input list:
▶ Expected return, eg estimated as the historical average return
▶ Risk or volatility, eg estimated as the standard deviation of
historical returns
▶ note: we need the full variance-covariance matrix!

▶ Optimal weights can be achieved in two steps!

1 / 34
Review — Two-fund separation

Optimal weights can be broken down in two steps!

1. Find out the mean-variance frontier from risky securities
▶ for each target return, the optimal mix that delivers least
variance
▶ find the tangency portfolio (the one with the highest Sharpe
ratio!)
▶ note: this portfolio does not depend on risk preferences (same
for all investors)

2. Mix this “efficient portfolio” with the riskfree asset based on

risk preferences
▶ appropriate mix of the tangency portflio and risk-free asset
▶ note: this step is investor-specific

2 / 34
PORTFOLIO CHOICE: CHALLENGES
MPT in Practice, Issues

▶ Issues with inputs!

1. large input set of expected returns, variances and covariances, if a
large number of securities, all estimated with error
2. need long history for expected returns, and variances and
covariances, and they also vary a lot over time!
▶ answer for next lecture: reduce dimension with factor models prelude

▶ Issues with outputs:

1. portfolio weights very sensitive to small changes in inputs, but
re-balancing needs large transaction costs
2. unconstrained optimization often leads to portfolios with large short
positions, levered positions, etc, not practical
▶ (partial answer: impose constraints on portfolio weights)

3 / 34
Alternative Approach

Solution: Skip doing the first step!

▶ Start with tradable risky portfolios. Index Funds. ETFs
▶ Already diversified mixed of individual securities...
✓ close enough to the tangency portfolio!

→ The rise of “Passive investment strategies!”

▶ Justification: CAPM

4 / 34
Example — Portfolio Choice with a Debt and Equity fund

40:60 weights on debt:equity (in this example) gives the best mix!
Why?

5 / 34
Example — Portfolio Choice with a Debt and Equity fund
Input list

Debt Equity
Expected return 8% 13%
Standard deviation 12% 20%
correlation .3

recall the formulas from statistics [w be the weight on Equity]:

r¯p = (1 − w )¯
rD + w r¯E

σp2 = (1 − w )2 σD
2
+ 2w (1 − w ) σD σP ρDE +w 2 σE2
| {z }
CovDE

rP − rf )/σP )
riskfree asset rf = 5% (to measure Sharpe Ratio = (¯

6 / 34
Example — Portfolio Choice with a Debt and Equity fund
The best risky mix!

portfolio P wdebt wequity r¯P σP Sharpe Ratio

100:0 1 0 8% 12% .25
80:20 .8 .2 9% 11.45% .349
60:40 .6 .4 10% 12.26% .408
40:60 .4 .6 11% 14.20% .423*
20:100 .2 .8 12% 16.88% .415
0:100 0 1 13% 20% .4

▶ The optimal risky portfo is 40:60, debt and equity funds!

▶ It gives the expected return 11%, STD 14.20%, SR = .423
▶ Quiz: re-do this with rf = 1%, r¯D = 4%, r¯P = 9% and show
optimal weights are the same
analytical derivations mixing with the riskfree asset, example

7 / 34
What makes this tractable?

▶ We do not need to solve for 500+ unknow portfolio weights

on individual stocks in the market...
▶ Just buy a fund in the market (Index ETFs)
▶ ...Such funds are already a good mix of individual assets
▶ ...Not much compromise on risk-return tradeoff
▶ Is this the case?
→ CAPM

8 / 34
CAPM
Capital Asset Pricing Model (CAPM)

▶ Developed by William Sharpe (published

paper: 1964, Nobel: 1990) and others

▶ Two key insights/results:

1. In equilibrium, the tangency portfolio is the market
portfolio
— implications for investment, portfolio choice
2. Specifies the expected return on a security and its risk
as measured by its beta. (wrt market portfo)
— asset pricing (our course: performance evaluation)

9 / 34
CAPM — Tangency Portfolio in Equilibrium
▶ Assumptions:
▶ investors are risk averse
▶ investors care only about mean and standard deviation or
security returns are normally distributed
▶ NEW: investors have the same estimates of expected returns,
volatilities and correlations (“homogeneous expectations”)
▶ NEW: investors can borrow/lend at the same riskfree rate
▶ Result: Investors will all identify the same tangency
portfolio (implied by rationality!)
▶ Econ 101: SUPPLY = DEMAND →

Tangency Portfo = Market

Verbal proof: If all investors demand the same portfolio and the
supply of risky securities is the “market portfolio” then, in
equilibrium, the tangency portfolio is the market portfolio

10 / 34
What is the market portfolio?

In theory:
▶ all risky securities in the economy (equities, bonds, gold, real
estate, labour income...)
▶ weighted by their market value

Market Value i
w i = PN
i=1 Market Value i

In practice, used proxies often consider equities only

▶ use broad value-weighted stock market indices eg S&P500, all
US stocks, all global stocks, etc

11 / 34
Why is the market portfolio tangent?

▶ Same beliefs =⇒ agree on tangency portfolio (PT )

▶ Everybody holds risky assets in same proportions
▶ Because risky holdings are proportional to the tangency
portfolio (from two-fund separation).
▶ I hold 20% risk-free bond, 80% tangency port
▶ You hold -30% risk-free bond, 130% tangency port
▶ Thus, sum of everybody’s holdings has these properties:
▶ Proportional to tangent portfolio (by two-fund sep.).
▶ Equals the market portfolio (by definition).
⇒ The market portfolio is proportionate to tangency portfolio
∗ We don’t need rationality and homogeneous beliefs for all
▶ random mistakes around the “correct” mean are just fine

12 / 34
Capital Market Line
▶ The Capital Allocation Line with the highest Sharpe Ratio now is
the Capital Market Like.
▶ The Capital Market Line (CML) is the CAL that combines
investments in the riskfree asset and the market portfolio
▶ Portfolio choice, 2nd step: mix riskfree and the market portfolio
[move along CML]

13 / 34
Expected Return on Individual Securities via CAPM
Review from Finance II
▶ What is the relationship between expected return and risk for
individual securities or portfolios other than the market
portfolio?
▶ What is the relevant measure of risk for an individual security
held as part of the market portfolio?
▶ Not volatility (standard deviation), as firm-specific risk is
diversified away, only market risk is priced
▶ Instead, the appropriate measure will be a function of how
much that security contributes to the risk of the market
portfolio, which in turn depends on the covariance of the
security’s returns with the market portfolio

cov (ri , rm ) σi
E [ri ] = rf + βi [E (rm ) − rf ] βi := 2
= ρi,m
σm σm

14 / 34
CAPM — Derivation, key idea!
Review from Finance II

▶ The CAPM is derived by noting that, in equilibrium:

▶ All securities have to earn the same expected excess return per
unit of risk contributed to the market portfolio...
▶ ...otherwise you can improve the risk-return characteristics of
the market portfolio by rebalancing portfolio weights

CAPM - Derivation [review from Finance II]

15 / 34
Interpreting Risk with CAPM
Review from Finance II

cov (ri , rm ) σi
E [ri ] = rf + βi [E (rm ) − rf ] , βi = 2
= ρi,m
σm σm
In words:
▶ Expected return on a share of stock =
Risk-free rate + Compensation for taking on risk
▶ Compensation for taking on risk =
Measure of risk (Beta) × Price of risk
▶ What is a priced risk?
▶ the one that exposure to it gives you extra expected return
▶ market, or systematic, risks
▶ so here, market exposure (measured by CAPM β) is priced.
▶ Beta captures the extent to which a share contributes to
portfolio risk
▶ high beta = high risk asset
▶ low beta = low risk asset
some review examples
16 / 34
Capital Market Line + Priced Risk of Individual Stocks
Review from Finance II
▶ No relationship: individual stock’s volatility and expected returns
▶ Stock’s expected return is due only to the fraction of its volatility
that is common with the market: ρi,m × σi
▶ The relationship between risk and return for individual stocks
becomes evident only when we measure “market” risk

17 / 34
Security Market Line (SML)
Review from Finance II
▶ The CAPM equation implies that there is a linear relationship
between a stock’s beta and its expected return:
E (ri ) = rf + βi · (E (rm ) − rf )
▶ The Security Market Line (SML) is the plot of the linear
relationship between expected return and beta from CAPM

18 / 34
Asset Pricing via CAPM: From Theory to Practice
Review from Finance II

cov (ri , rm )
E [ri ] − rf = βi [E (rm ) − rf ] , βi = 2
σm
▶ Risk-free rate rf ? Easy to look up, for a safe asset
▶ Use the yield on short-term government debt.
▶ Need to estimate:
▶ Beta: βi
▶ Price of risk: rm − rf
▶ How we estimate β? Typically from historical returns data!
1. Based directly on estimates of variances and covariances
σi,m
βi = 2
σm
2. Using OLS regression analysis:
▶ Run following regression: ri − rf = ai + bi (rm − rf ) + εi
▶ regr coeff bi is the estimate of βi (seminar 1, also in the
coming lectures) estimating beta, details + examples

19 / 34
CAPM, Summary
▶ An equilibrium model, describing the relationship between risk
and return, while characterizing the efficient portfolio
▶ Principal results:
1. The market portfolio is the tangency portfolio — it has
the highest Sharpe Ratio
2. The market portfolio is an efficient portfolio — all
investors will hold a combination of the market and the
risk-free asset.
3. The risk premium on an asset will be proportional to the
market risk premium and to the asset’s beta wrt market portfo
▶ Estimating equity cost of capital for company valuation - see
Corporate Finance course
▶ Measuring performance, while controlling for the risk
exposure, in the asset management industry
▶ In “event studies”’ to measure how company’s share prices react to
specific events/corporate announcements, etc. - see Corporate
Finance course

20 / 34
PASSIVE INVESTMENT STRATEGIES
Active vs Passive Investment Strategies

▶ Passive: goal is to track an index or benchmark as closely as

possible; minimal portfolio re-balancing
▶ Active: goal is outperform the benchmark using a variety of
techniques; stock picking, market timing, etc

▶ The rise of passive strategies with index funds...

Fidelity CEO, Abigail Johnson, on Active vs. Passive Investing:

https://www.youtube.com/watch?v=PDctKQPXN7Q

21 / 34
The rise of Passive Investment

https://www.economist.com/briefing/2014/05/01/will-invest-for-food

▶ realized in 2020: active $74T, passive $23T

▶ projection 2025: active $88T, passive $37T
https://www.pwc.com/jg/en/publications/asset-and-wealth-management-revolution.html#download
22 / 34
The rise of Passive Investment
▶ inflows and outflows:

https://www.morningstar.com/business/insights/blog/funds/
active-vs-passive-investing

23 / 34
Passive Funds — example
“Storebrand Sverige (formerly SPP Aktiefond Sverige) is an
index-based equity fund...”

https://www.morningstar.se/se/funds/snapshot/snapshot.aspx?id=F0GBR04FXQ

24 / 34
Passive Funds — example

https://www.morningstar.se/se/funds/snapshot/snapshot.aspx?id=F0GBR04FXQ

25 / 34
Active Funds — example
“Swedbank Robur Sweden is an actively managed equity
fund....”

https://www.morningstar.se/se/funds/snapshot/snapshot.aspx?id=F0GBR04LSI

26 / 34
Active Funds — example

https://www.morningstar.se/se/funds/snapshot/snapshot.aspx?id=F0GBR04LSI

27 / 34
The Passive Investment Strategy

▶ Passive strategy: a portfolio decision that avoids any direct or

indirect security analysis.
▶ The natural candidate for a passive strategy: a well-diversified
portfolio of common stocks. “the entire market”
▶ Hold stocks proportionate to their market value — a
value-weighted index! Index Funds
▶ Mix this with a riskfree asset — government notes, money
market funds, etc.
▶ Example: T-bills + Vanguard/500 Index Fund (tracks US
S&P 500), Avanza Zero (tracks Sweden SIX30RX)
https://investor.vanguard.com/investment-products/
mutual-funds/profile/vfiax

28 / 34
The Passive Investment Strategy
Advantages

✓ Cost-efficient: no time or cost or talent or a PhD in Finance

(!) to acquire information and analyze the risky assets as in
an active portfolio management. easy to implement...
∗ little management fees charged by passive funds...
✓ Free-rider benefits: professional investors/skilled acitve
managers push prices to their “fair values” (fundamentals of
risk and return)
✓ Already well-diversified! (if anything, the gain in the Sharpe
Ratio would be minimal under alternative strategies)
✓ Justified by theory and evidence!

29 / 34
CAPM =⇒ “The Passive Strategy (=market) is Efficient”

The justification for Passive Investment Strategies:

▶ recall — CAPM: the tangency/efficient portfolio is the
“market” portfolio
▶ new interpretation: the market portfo summarizes all the
info on the universe of securities that you need to know for
the portfolio choice
▶ with ETFs and Index Fund you implement it with minimal
trading/transaction costs
→ “Mutual fund theorem”

Warren Buffett on passive index investing vs. active money managers (2020):
https://youtu.be/9BOefaL-RPI?si=QAQfYH_fkb6omsSi

30 / 34
Inferior Performance of Actively Managed Funds
50 years of data

Wilshire 5000 average return = 12.49%

Actively managed equity funds average return = 11.53%

31 / 34
Active vs Passive Performance

32 / 34
Active vs Passive Performance, European Funds

https://www.ft.com/content/e555d83a-ed28-11e5-888e-2eadd5fbc4a4

33 / 34
Any many more notes and reports on this...

https:
//
www.
ft.
com/
content/
e139d940-977d-11e6-a1dc-bdf38d484582

34 / 34
Where are we?

“Basics of Portfolio Choice”

Today’s Lecture:
▶ Portfolio choice with funds, Example: Equity/Debt Funds
▶ CAPM — implications for the portfolio choice
▶ Passive investment strategies (Index Funds)

Next Lecture:
▶ Active allocations (via Single-Index Models)
Key Terms

▶ homogeneous expectations
▶ market portfolio
▶ capital market line (CML)
▶ capital asset pricing model (CAPM)
▶ market price of risk
▶ beta
▶ security market line (SML)
▶ active investment strategy
▶ passive investment strategy
▶ exchange traded funds
▶ index funds
APPENDIX
Example: Factor model return correlations
Example: Single Factor Model, return correlations
▶ Consider the following model of returns for securities:

R1 = α1 + β1 Rm + ε1

R2 = α2 + β2 Rm + ε2
..
.
▶ in which αs are constants and ε are noise with E(ε) = 0
▶ Stocks’ returns depend on a common (random) factor Rm
▶ Assume that the errors are uncorrelated with each other and
with the common factor Rm , i.e.,
Cov (εi , εj ) = 0, Cov (εi , Rm ) = 0, Cov (εj , Rm ) = 0
▶ From this, it follows that Cov (Ri , Rj ) = βi βj Var (Rm )
▶ Also, note that Cov (Ri , Rm ) = βi Var (Rm )
▶ We only need 3N inputs for this model (instead of N 2 )

go back
Example: Single Factor Model, CAPM

▶ Exposure to the market portfolio: Covi,m /Varm = βi

▶ Expected return according to CAPM:
E [Ri ] = E [ri ] − rf = βi (E [rm ] − rf ) = βi E [Rm ]
▶ Expected return from the specified relationship:
E [Ri ] = αi + βi E [Rm ]
▶ Hence,
CAPM ⇐⇒ αi = 0, ∀i

go back
Mean and Variance of a Portfolio of Assets
Mean and Variance of a Portfolio of Assets

▶
n
X
E [rp ] = wi E [ri ]
i=1

▶
n X
X n
σp2 = wi wj Cov (ri , rj )
i=1 j=1

▶ wi s are portfolio weights

Tobin step 1: 2 Asset case
Tobin step 1: 2 Asset case

▶ To compute the mean-variance efficient (MVE), or tangency

portfolio, maximize the Sharpe ratio:
E (rp )−rf
max σp
w
where E (rp ) = wE (rA ) + (1 − w)E (rB )
1/2
σp = w2 σA2 + (1 − w)2 σB 2 + 2w(1 − w)ρ

A,B σA σB

▶ The solution to this is ugly

go back
Solution of the MVE problem

▶ The tangency portfolio weight for A:

2 − E (R ) ρ
E (RA ) σB σ σ
wT =
2

2
 B
A,B A B

E RA σB + E RB σA − E RA + E RB ρA,B σA σB

▶ Where R denotes the excess return

E (Ri ) = E (ri ) − rf i = A, B

▶ (sometimes we show it also with the notation rie )

go back
Another Numerical Example, Input list

Asset Expected Return Volatility Correlation

Asset A Asset B Risk-free Asset
Asset A 10% 20% 1 0.5 0
Asset B 15% 30% 0.5 1 0
Risk-free Asset 3% 0% 0 0 1
Minimum variance 10.75% 19.64% wA = 0.85
(only risky assets) we already saw how to derive this portfolio.

go back
Another Numerical Example, Optimal weights

▶ Tangency portfolio weight on asset A:

▶ How much you should put in B?

go back
Another Numerical Example, The tangency portfolio

Given the weights, we can compute the risk and return of the
tangency portfolio:

E (rT ) = wT E (rA ) + (1 − wT ) E (rB )

= 0.5 · 0.10 + 0.5 · 0.15 = 0.125 = 12.5%
1/2
σT = w σA + (1 − w)2 σB
 2 2 2
+ 2w(1 − w)ρA,B σA σB = 21.79%

go back
Another Numerical Example, Maximal Sharpe Ratio?

▶ This portfolio should have the maximal Sharpe ratio

▶ Does it? Compute Sharpe Ratios

go back
Another Numerical Example, The Sharpe Ratio

▶ Sharpe ratios of the stocks A and B

E(rA )−rf 0.10−0.03
SRA = σA = 0.20 = 0.35
E(rB )−rf 0.15−0.03
SRB = σB = 0.30 = 0.40

▶ Sharpe ratio of the tangency portfolio

E (rT ) − rf 0.125 − 0.03

SRT = = = 0.4359
σT 0.2179

go back
 Portfolio Choice,
Mixing with the riskfree Asset
Portfolio Choice with a Debt and Equity fund
Mixing with the riskfree!

▶ Let’s say, achieve a target rate 9%.

▶ Solve for wP using 9% = (1 − wP ) ∗ rf + wP ∗ r¯P
▶ 9 = 5 + wP ∗ 6, so wP = 4/6 = 2/3 = 67%
▶ Put one-third in the risk-free rate. two-thirds in the tangency
portfolio P
▶ And we now wP loads on debt:equity by 40:60
▶ Complete portfolio: riskfree = 1/3 = 33%, debt fund =
2/3*4/10 = 27%, equity fund = 2/3*6/10 = 40%
▶ note: the relative weight of debt:equity will always be 40:60,
regardless of the target rate!
▶ The corresponding vol: σQ = wP σP = .67 ∗ .14.20% = 9.47%
— check that SRQ = .423 like SRP

go back
CAPM - Derivation [review from Finance II]
CAPM - Derivation [review from Finance II]

▶ For security i held with weight wi as part of the market

portfolio:
▶ Contribution to excess expected return of market
portfolio:
wi (E [ri ] − rf )
▶ Contribution to variance of market portfolio:

wi Covi,m

go back
CAPM - Derivation [review from Finance II]
Therefore, for security i:
▶ contribution to excess expected return of market portfolio
contribution to excess expected return of market portfolio E [ri ] − rf
=
contribution to variance of market portfolio Covim
▶ In equilibrium this has to be the same for all securities
(otherwise you can imporove the risk-return properties of the
market portfolio) and it has to equal the ratio for the market
as a whole:
E [ri ] − rf E [rj ] − rf E [rm ] − rf
= = 2
∀i, j
Covim Covjm σm
▶ Rearranging gives the CAPM:
Covi,m
E [ri ] − rf = 2
(E [rm ] − rf )
σm

go back
CAPM - Derivation [review from Finance II]

Covi,m σi
E (ri ) = rf + βi · (E (rm ) − rf ) where βi = 2
= ρi,m
| {z } σm σm
risk premium for security i

What it does/says:
▶ gives the relationship between expected return and risk (as
measured by beta) that holds for all individual securities,
portfolios of securities*
▶ the expected return on any security i is equal to:
▶ the risk free rate plus
▶ a risk premium that is equal to the security’s beta multiplied
by the market risk premium (excess expected return)
∗ The beta of a portfolio is simply βP = N
P
PN i=1 wi · βi where
i=1 wi = 1. The beta of the market portfolio is 1.

go back
CAPM - Examples
We measure CAPM risk with beta

▶ Beta captures the extent to which a share contributes to

portfolio risk
▶ High beta = high risk
▶ Low beta = low risk
▶ Intuitively, beta is a measure of a share’s cyclicality
▶ Beta tells you how much the stock return moves for x%
change in the market return.
▶ If beta = 1.5, then 1% change in market leads to 1.5% change
in the stock return (on average!)
▶ If beta = -0.5, then 1% change in market leads to -0.5%
change in the stock return (on average!) — hedging against
the market risk

go back
CAPM - Computing the expected return for a Stock

Example: Suppose the risk-free return is 4% and the market

portfolio has an expected return of 10% and a volatility of 16%.
Apple stock has a 22% volatility and a correlation with the market
of 0.50.
▶ What is Apple’s beta with the market?
Answer:
σi 22%
β = ρi,m × = 0.50 × = 0.69
σm 16%
That is for each 1% move of the market portfolio, Apple stock
tends to move 0.69%.

go back
CAPM - Computing the expected return for negative-beta
Stock
Example: Suppose the risk-free return is 4% and the market
portfolio has an expected return of 10%. Moreover, suppose stock
A, has a negative beta of -0.30.
▶ How does its expected return compare to the risk-free rate,
according to the CAPM?
▶ Does this result make sense?
Answer: If CAPM holds, then

E (rA ) = rf +β×(E (rm ) − rf ) = 4%−0.3×(10% − 4%) = 2.2 < 4%

Stock A provides “recession insurance” for the portfolio. That is,

when times are bad and most stocks are down, stock A will do well
and offset some of this negative return. Investors are willing to pay
for this insurance by accepting an expected return below the
risk-free rate.
go back
Final exam type of question

Question (5 points): Suppose that the expected return on the

market is 12%, the risk free rate is 5%, and the standard deviation
on the market is 20%.
1. According to the CAPM, what is the β on a portfolio that
achieves an expected return of 16%?
2. Now assume this portfolio is efficient. What must the
standard deviation on this portfolio be?
3. What is the relation between β and standard deviation for
efficient portfolios? Does this relation hold for inefficient
portfolios?

go back
Final exam type of question
Question (10 points): Assume the CAPM holds. You are given
the following information on returns on efficient portoflios:
Mean Standard Deviation
Portfolio 1 0.0982 0.12
Portfolio 2 0.2270 0.35

1. Calculate the return on the risk free rate.

2. Calculate the Sharpe ratio on the market portfolio.
3. Consider an inefficient portfolio with a standard deviation equal to
that of Portfolio 1. Will the expected return be higher or lower than
the expected return on Portfolio 1, or can you not tell? Explain your
answer.
4. Consider the same inefficient portfolio as in (c). Will the β on this
portfolio be higher or lower than the β on Portfolio 1, or can you
not tell? Explain your answer.

go back
CAPM - How we estimate β?
Estimation choices

Different services come up with different betas, because they use

different approaches to measure co-movement with the market:
▶ How many years of data (time horizon)?
▶ What periodicity?
▶ Which market?
Recommendations:
▶ 2-5 years (2 years if firm is changing, 5 if stable).
▶ Weekly is best, monthly second best.
▶ Use the market relevant to your investors.
▶ Consider ”Bloomberg Adjustment” Recommendations:
▶ 2/3 × data beta + 1/3 × 1 = adjusted beta

go back
Estimating Beta - Example

▶ Summary stats on excess returns:

σi,m ρi,m σi
βi = 2
= 1.2608 =
σm σm
▶ Estimate beta based on covariance and variance data ↓

go back
Estimating Beta - Example
▶ Alternatively, using OLS, find best-fitting∗ straight line for the
data:
▶ Slope of this line is the estimate of beta

∗ Thatminimises sum of squared errors between each data point

and the regression line in the y dimension
go back
Estimating Beta - Example

▶ Run the regression:

ri − rf = ai + bi (rm − rf ) + ϵi

▶ bi – slope of the regression line and the estimate for beta

▶ ai – intercept term and should be zero if the CAPM holds
▶ ϵi is the error or residual term

go back
Estimating Beta - Example
▶ Regression results:

▶ For each coefficient, we have a p-value (and t-stat):

▶ p-value: probability of observing the estimated, or a more
extreme, value if the true value of the parameter is zero
▶ typically require p-value less than or equal to 0.05 (or a t-stat
greater than 2) for the coefficient to be statistically significant
(statistically significantly different from zero)

go back
Estimating Beta - Example

▶ Regression results:

▶ Output also includes the Regression R 2 :

▶ how much of the variance of the dependent variable (here
ADBEs excess returns) is explained by the independent
variable (here S&P500 excess returns)
▶ also correlation (ρi,m ) squared

go back
Estimating Beta - Example
Data with OLS regression line:

go back
Regressions in Excel

Excel
▶ Make sure that the regression module is installed
(File =⇒ Options =⇒ Add-Ins =⇒ Analysis Toolpack)
▶ Implementation
(Data =⇒ Data Analysis =⇒ Regression)
I recommend you using R: http://cran.r-project.org/
▶ Steeper learning curve, but more powerful

go back
OLS and Risk Decomposition
▶ OLS analysis allows you to easily decompose the risk of the
stock:
▶ Start with the OLS regression equation:

ri − rf = ai + bi (rm − rf ) + ϵi

▶ If you assume a constant riskfree rate then, by the properties

of OLS, specifically corr(rm , ϵi )=0 :

go back
Estimating Beta - Bloomberg (BETA)
Using 5 years of monthly data:

go back
Estimating Beta - Bloomberg (BETA)
Using 5 years of weekly data:

go back