Beta Estimation & Single Index Model

Prof(Dr.) Harsh Vardhan

There are two times in a man’s life
when one shouldn’t speculate. When
he can’t afford it and when he can.
-Mark Twain

CF/MDI/Harsh/12 2
 Objectives

▪ CAPM Review
▪ Discuss The Methods Of Estimating Beta.
▪ Explain the Market Model for Calculating Beta.
▪ Single Index Model
▪ Examine the difference between Betas of individual firms and the
Industry Beta.
▪ Highlight the Beta instability.
▪ Usage of Beta in determining the Cost of Equity.

CF/MDI/Harsh/12 3
 CAPM-Review
The Capital Asset Pricing Model (CAPM). E(Ri) = Rf+ βim [E(Rm) - Rf] ; where:
▪ Expected Return on the capital asset E(Ri)
▪ Risk-free rate of interest Rf
▪ Beta coefficient, the sensitivity of the asset returns to market returns,
𝐶𝑂𝑉 (Ri ;Rm)
▪ βim = 2
,,un-diversifiable risk of security i,
σ𝑚
σ𝑖
▪ βim = ρim * ,where ρim is correlation co-efficient between E(Ri) & E(Rm)
σ𝑚
▪ Expected Return of the market E(Rm)
▪ Market Premium or Risk Premium (the difference between the expected market rate
of return and the risk-free rate of return) [E(Rm)-Rf]

CF/MDI/Harsh/12 4
 The Capital Asset Pricing Model

Expected E(R) = RF+ β [E(RM) - RF]

Return E(R) – RF = β [E(RM) - RF]
E(R)
E(RM)

E ( R ) − RF =  [ E ( RM ) − RF ]
RF

Beta
1.0

CF/MDI/Harsh/12 5
 Capital Market Line
▪ Portfolio M is optimum risk portfolio, which can be combined with risk free
asset
▪ The slope of CML describes the best price of a given level of risk in equilibrium.
The slope is also known as risk to reward ratio.
 E ( Rm ) − R f 
Slope of CML =  
  m 
▪ The expected return on a portfolio on CML is defined by the following equation:

 E ( Rm ) − R f 
E ( Rp ) = R f +  p
 m 

CF/MDI/Harsh/12 6
 Capital Market Line
▪ Capital Market Line-The expected return on a portfolio on CML is
defined by the following equation:
 E ( Rm ) − R f 
E ( Rp ) = R f +  p
 m 
Where,
▪ E(Rp)= Expected return on portfolio along the Capital Market Line.
▪ Rf =Risk free return
▪ E(Rm)=Expected Market Portfolio Return
▪ σm =Market Portfolio Standard Deviation
▪ σp =Standard Deviation of Portfolio along CML.

CF/MDI/Harsh/12 7
 Security Market Line (SML)
▪ For a given amount of systematic risk (), SML
shows the required rate of return.

E(Rj)

βj =
𝐶𝑂𝑉 Rj;Rm , , …… 1 E(R j ) = R f +  (R m ) – R f  β j
σ2𝑚 SLM

𝐶𝑂𝑉(Rj;Rm)
ρjm = ,….. 2
σ𝑗 σ𝑚 Rm

Put value of COV Rj;Rm

from equation 2 in Rf

equation 1
 = (covarj,m/2m)
σ𝑗 σ𝑚 ∗ρjm 0 1.0
βj = 2 σ𝑚
σ𝑗
βj = * ρjm …..3 CF/MDI/Harsh/12 8
σ𝑚
 Beta Estimation
Direct Method
The ratio of covariance between market return and the security’s
return to the market return variance:
Covar j, m
j =
σ 2m
σ j σ m Cor j, m σj
= =  Cor j, m
σm  σm σm
Where βj is beta of security j
COV Rj ;Rm is Covariance between security j & security m.
σ𝑗 and σ𝑚 are standard deviations of security j & security m.

9 CF/MDI/Harsh/12
 Example
The table below provides % returns on the market represented by
Sensex (Sensitivity Index )and Jaya Infotech for last five years:
Determine Beta of Jaya Infotech & equation of its Characteristic Line

Returns on Sensex and Jaya Infotech

10d
CF/MDI/Harsh/12
 Example

Beta Calculation for Jaya Infotech Limited

11 CF/MDI/Harsh/12
 Steps to be followed

12 CF/MDI/Harsh/12
 Cont…

13 CF/MDI/Harsh/12
 Cont…

14 CF/MDI/Harsh/12
 Linear Regression

▪ Linear Regression estimates coefficients of the linear equation, involving one

or more independent variables, that best predicts the value of the
dependent variable.
▪ For example, you can predict a salesperson's total yearly sales (the
dependent variable) from independent variables such as age, education, and
years of experience.

CF/MDI/Harsh/12 15
 Simple Linear Regression Model-Recap

▪ Relationship Between Variables is a Linear Function

▪ The Straight Line that Best Fit the Data

Y intercept Random
Error

Yi =  0 +  1 X i +  i
Dependent
(Response) Independent
Slope (Explanatory)
Variable
Variable
CF/MDI/Harsh/12 16
 Population Linear Regression Model

Y Yi =  0 + 1X i +  i Observed
Value

i = Random Error

m =  0 + 1X i
YX

X
Observed Value
CF/MDI/Harsh/12 17
 Measures of Variation: The Sum of Squares

Y 
SSE =(Yi - Yi )2
_
SST = (Yi - Y)2

 _
SSR = (Yi - Y)2
_
Y

X
Xi
CF/MDI/Harsh/12 18
 The Coefficient of Determination

SSR Regression Sum Of Squares

r2 = =
SST Total Sum Of Squares

▪ The Coefficient of Determination is equal to the Regression

Sum of Squares (explained variation) divided by the Total
Sum of Square (total variation) .
▪ The Coefficient of Determination measures the proportion of
variation in Y that is explained by the independent variable X
in the Regression Model CF/MDI/Harsh/12 19
 Standard Error of Estimate

n

S yx =
SSE  i
(Y − Yi ) 2

n−2 = i =1
n−2

The Standard Deviation of the variation of

observations around the regression line

CF/MDI/Harsh/12 20
 Inferences about the Slope: t- Test
▪ t- Test for a Population Slope
▪ Is there a Linear Relationship Between X & Y ?
▪ Null and Alternative Hypotheses
H0: 1 = 0 (No Linear Relationship)
H1: 1  0 (Linear Relationship)
SYX
Sb1 =
b1 − 1 n
▪ Test Statistic: t= Where
(X i − X) 2

Sb1 i =1

and df = n - 2
CF/MDI/Harsh/12 21
 Example : Determining Beta & intercept values

▪ Ms. A invested in the XYZ company some time ago and now wishes to
find out how her shares have been performing in relation to the
overall market. Ms. A has collected monthly % returns over the past
year on both shares(Ri) and Market index (Rm) as given in the table:
Market Stock
Month
returns (Rm) Returns (Ri)
1 7.98 9.33
2 4.55 6.72
3 1.04 -1.21
4 4.42 0.86
5 5.76 1.34
6 -1.67 -4.73
7 4.18 7.56
8 6.7 5.39
9 2.87 2.4
10 -3.89 -7.9
11 -1.3 -1.7
12 2.08
CF/MDI/Harsh/12 6.58 22
 Month Market returns Rm Stock Returns,Ri
1 7.98 9.33
2 4.55 6.72
3 1.04 -1.21
4 4.42 0.86
5 5.76 1.34
6 -1.67 -4.73
7 4.18 7.56
8 6.7 5.39
9 2.87 2.4
10 -3.89 -7.9
11 -1.3 -1.7
12 2.08 6.58
-1.3143 α intercept
1.2351 β slope

Market Variance σ2m 13.0453

R 0.8423
R2 0.7095 Systematic Risk
R 2 is a measure 71% variabilty in returns of company
of systematic risk can be expained by the variablity
in the market Index Rm
CF/MDI/Harsh/12 23
 Systematic & Unsystematic Risk
Month Market returns Rm Stock Returns,Ri
1 7.98 9.33
2 4.55 6.72
3 1.04 -1.21
4 4.42 0.86
5 5.76 1.34
6 -1.67 -4.73
7 4.18 7.56
8 6.7 5.39
9 2.87 2.4
10 -3.89 -7.9
11 -1.3 -1.7
12 2.08 6.58
INTERCEPT(C3:C14,B3:B14) -1.3143 α intercept
SLOPE(C3:C14,B3:B14) 1.2351 β slope SLOPE(C3:C14,B3:B14)

Market Variance σ2m 13.0453 VAR(B3:B14)

Y=α +βX
R 0.8423 CORREL(B3:B14,C3:C14)
R2 0.7095 Systematic Risk
R 2 is a measure 71% variabilty in returns of company
of systematic risk can be expained by the variablity
in the market Index Rm
(1- R2) 0.2905 Unsystematic Risk

CF/MDI/Harsh/12 24
 SIM Model
12

10

2
Stock Return Ri Linear (SIM Model)
0
-6 -4 -2 0 2 4 6 8 10
-2

-4
y= -1.3153 + 1.237 x
-6 R2=0.7095

-8

-10
Market Index Return Rm

CF/MDI/Harsh/12 25
 SUMMARY OUTPUT

Regression Statistics
Multiple R 0.842290118
R Square 0.709452643
Adjusted R
Square 0.680397907
Standard Error 2.994121758
Observations 12

Coefficients Standard Error t Stat P-value

Intercept -1.314346637 1.100695717 -1.19411 0.259993
X Variable 1 1.235090454 0.249945616 4.941437 0.000586

Y= -1.3153 + 1.237 X
P-value< 1%= Reject H0

CF/MDI/Harsh/12 26
 Interpretation
▪ β= 1.2351 and intercept α = - 0.0131.
▪ A security with β=1 moves with market i.e it follows the general trend of
all securities.
▪ β = 1.2351 means the stock is more volatile than overall market and will
have relatively higher return.
▪ The Coefficient of Determination R2 measure goodness of fit of
regression line. If the fit is good, R2 =1 ,the fit is poor R2 is close to 0.
▪ R2 =0.7095 which indicate a good line fit. R2 is also a measure of
Systematic risk i.e 70.95% of the of the totality in company’s returns can
be explained by the variability in the market index returns, Rm.
▪ The remaining % , (1-R2 )=29.05% is attributed to Unsystematic risk.

CF/MDI/Harsh/12 27
 Betas for some Sensex Companies -Historical data
▪ The BSE’s sensitivity index includes 30 highly traded shares.
▪ The estimates are based on daily returns for one year.

Note that Jaiprakash Associates

has the highest beta of 2.28 and
Gujarat Ambuja Cement the lowest
28 beta of 0.37.
CF/MDI/Harsh/12
 Does Beta Remain Stable Over Time?

▪ Betas may not remain stable for a company over time even
if a company stays in the same industry.
▪ Over time, a company may witness changes in its product
mix, technology, competition or market share.

29 CF/MDI/Harsh/12
To achieve superior performance , you have to be
different from majority.

CF/MDI/Harsh/12 30
 Market Model
▪ The return on a share and market index may be calculated as total
return that is ,dividend yield plus capital gain.
▪ The Market model is given by the following regression equation:
▪ Rj= α +βjRm +ej
▪ The slope β of the regression measures the variability of the
security’s return relative to the market return and it is the security’s
beta.
▪ The compounded rate of return may be calculated by:
Pt
▪ rj= ln[Pt-Pt-1]= ln[ ]>0
Pt−1

CF/MDI/Harsh/12 31
 Beta Estimation in Practice

▪ The return on a share and market index may be calculated

as total return; that is, dividend yield plus capital gain:

▪ One may calculate the compounded rate of return by:

Pt
rj= ln[Pt-Pt-1]= ln[ ]
Pt−1
32 CF/MDI/Harsh/12
 Beta Calculation: Example
Let Ry is return of stock Y and Rx is corresponding market return for 5 years.

Y= - 0.8915+0.8824 X
33 CF/MDI/Harsh/12
 Beta Calculation: Example
▪ Y=α+ β*X …..(1)
▪ As 𝑋ത & 𝑌ത passes through equation 1
▪ 𝑌ത =α + β *𝑋ത …..(2)
▪ Therefore α = 𝑌ത - β *𝑋ത ….(3)
(𝑛 σ 𝑋𝑌−σ 𝑋∗σ 𝑌)
β= σ 2 σ
……(4)[=COV(X,Y)/VAR(X)]
(𝑛 𝑋 −( 𝑋)2)
▪ From equation (4), β =0.8824
ത 5.48 , 𝑌ത =3.95 and β =0.8824 in equation 3
▪ Put 𝑋=
▪ Thus α = - 0.8915,
▪ Putting value of α & β in regression equation (1)
▪ The estimated regression equation is:
▪ Y= - 0.8915+0.8824X.
▪ This implies if there is 1% change in the Market index then there is 0.88%
corresponding change in the stock value . Therefore the stock is defensive.
CF/MDI/Harsh/12 34
Single Index Model

CF/MDI/Harsh/12 35
No stock is inherently good or bad,
it’s the price that makes it so.

CF/MDI/Harsh/12 36
 Single Index Model

▪ The Single Index Model (SIM) ,developed by William Sharpe , is a simplified

version of the Markowitz Model.
▪ SIM’s main benefit is that it greatly reduces the amount of computation.
▪ A SIM portfolio analysis on n -securities requires only (n+1) calculations to build
n(n+1)
Variance -Covariance Matrix where as Markowitz Model generates
2
calculations to calculate similar n security analysis.
▪ For example a portfolio containing 100 securities needs only 101 calculation
using SIM but 5,050 calculations for the Markowitz Model.

CF/MDI/Harsh/12 37
▪ The SIM model assumes that there is a linear relationship between the price
movement of a security (or portfolio) and the price movement of the overall
market.
▪ The market is represented by a market index such as Sensex (India),
FTSE(UK), Dow Jones(USA) , Nikkei(Japan).
▪ The basic SIM straight line equation is: Ri= αi +βiRm +εi where i= 1,2,3….n
▪ Ri = the rate of return for security i over specified period
▪ Rm = the rate of return of the market index over specified period
▪ αi = the rate of return that is independent of market movement
▪ βi = measures the expected change in Ri to a given change in Rm , is called
beta - coefficient or simply beta
▪ εi = the residual (i.e. error) associated with Ri
CF/MDI/Harsh/12 38
▪ The equation Ri= αi +βiRm +εi divides the return on a stock into two
components.
▪ The first part (βiRm ) is due to market change is called Systematic or Market
risk.
▪ A βi of 3 means that a stock’s return is predicted to increase (decrease) by
3% when market increase (decrease) by 1%.
▪ The second part (αi +εi ) ,which is independent of the market is known as
the Unsystematic or Diversifiable risk.
▪ This type of risk which is specific to the individual stock can be reduced or
even eliminated by creating a well-balanced portfolio of diverse securities -
hence Diversifiable risk.
▪ In coordinate geometry terms , βi represents the slope of a line with αi the
intercept on the vertical y-axis. The horizontal x-axis is represented by
market index Rm and the y -axis by stock return Ri

CF/MDI/Harsh/12 39
 Components of Risk
▪ Market or systematic risk: risk related to the macro
economic factor or market index.
▪ Unsystematic or firm specific risk: risk not related
to the macro factor or market index.
▪ Total Risk = Systematic + Unsystematic

CF/MDI/Harsh/12 40
 Security Characteristic Line
Excess Returns
SCL

. . . .. .
. . . . ..
. .. . . . . . .
. . .
. . . . . .
. .. . . . . Excess returns

. . . ..
on market index
.
. . . .. . . .
Ri =  i + ßiRm + ei
CF/MDI/Harsh/12 41
 Q1
▪ Calculate the expected rate of return for security j from following
information
▪ Rf=10%,Rm=18%,βj= 1.35
▪ E(Rj)= Rf +βj (Rm - Rf)=10%+1.35( 18%-10%)= 20.8%

CF/MDI/Harsh/12 42
 Q2
▪ P Ltd has expected return of 22% and standard deviation of 40%.Q
Ltd has an expected return of 24% and standard deviation of 38%.P
has a beta of 0.86 and Q’s beta is 1.24.The correlation between the
return of P and Q is 0.72.The standard deviation of the market return
in 20%.
▪ Is investing in Q better than investing in P?
▪ If you invest 30% in Q and 70% in P,what is your expected rate of return and
portfolio standard deviation?
▪ What is the market portfolio expected rate of return and how much is the
risk free rate?
▪ What is the beta of portfolio if P’s weight is 70% and Q’s 30%.

CF/MDI/Harsh/12 43
 Q3
▪ An aggressive mutual fund promises an expected
return of 16% with a possible volatility of 20%.On the
other hand ,a conservative mutual fund promises an
expected return of 13% and volatility of 15%
▪ Which fund would you like to invest in?
▪ Would you like to invest in both ?
▪ Assuming you can borrow money from your provident fund
at an opportunity cost of 10%,which fund you would invest
your money in?

CF/MDI/Harsh/12 44
Thanks

CF/MDI/Harsh/12 45