Scribd
Upload
76 views

WS Simple Regression Analysis Download

This document provides an overview of simple regression analysis using a single-index model. It defines the dependent and independent variables as well as the residual in a regression model. It then makes assumptions about the expected value, variance, and covariance of the residual. It explains how to measure the explanatory power of the regression using the coefficient of determination and how to estimate the slope from a sample using ordinary least squares. Finally, it provides a formula for calculating the covariance between two assets.

Uploaded by

Ajeet Yadav
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
76 views

WS Simple Regression Analysis Download

This document provides an overview of simple regression analysis using a single-index model. It defines the dependent and independent variables as well as the residual in a regression model. It then makes assumptions about the expected value, variance, and covariance of the residual. It explains how to measure the explanatory power of the regression using the coefficient of determination and how to estimate the slope from a sample using ordinary least squares. Finally, it provides a formula for calculating the covariance between two assets.

Uploaded by

Ajeet Yadav
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Math Workshop

Simple regression analysis


Example: Single-index Model

Regression model: Ri = αi + βi Rm + ei ;

Ri: dependent variable


Rm: independent variable
ei: residual

To go further, we make the following assumptions.


(1) E(ei)=0
(2) Var(ei)=σ2
(3) Cov(ei , ej)=0
(4) Cov(ei , Rm)=0

Var(Ri) = βi2 Var(Rm) + Var(ei) = βi2 σ2m + σ2

systematic risk firm-specific risk


(Explained variance) (Unexplained variance)

Also, Cov(Ri , Rm) = Cov(αi+βiRm+ei , Rm) = Cov(βiRm , Rm) = βiσ2m


 Slope βi = Cov(Ri ,Rm)/Var(Rm) = ρσ iσ m / σ m2 = ρσ i / σ m ........... (5)

Measure the explanatory power of the regression


 The coefficient of determination = βi2σ2m/Var(Ri) = ρ 2 (from (5))

Estimate the slope βi from a sample of observations (Ordinary Least Squares Estimator):

 βi =
∑ ( R − R )(R
i i m − Rm )
(from (5))
∑ (R − R
m m)
2

 αi = Ri − β̂ i Rm

Note: Cov(RA , RB) = Cov(αA+βARm+eA , αB+βBRm+eB)


= Cov(βARm , βBRm) = βA*βB*Var(Rm)

1
1. The market and stock J have the following probability distributions:
Probability rM rJ
0.3 15% 20%
0.4 9% 5%
0.3 18% 12%
(a) Calculate the expected return for the market and stock J.
(b) Calculate the standard deviations for the market and stock J.
(c) Calculate the beta for stock J.
(d) Calculate the correlation between stock J and the market.

Solution: (a) E(rM) = 0.3*15%+0.4*9%+0.3*18% = 13.5%


E(rj) = 0.3*20%+0.4*5%+0.3*12% = 11.6%

(b) Var(rM) = 0.3*(0.15-0.135)^2+0.4*(0.09-0.135)^2+0.3*(0.18-0.135)^2 = 0.001485


SD(rM) = SQRT(0.001485) = 3.85%
Var(rj) = 0.3*(0.20-0.116)^2+0.4*(0.05-0.116)^2+0.3*(0.12-0.116)^2 = 0.003864
SD(rj) = SQRT(0.003864) = 6.22%

(c) Betaj = Cov(Rj ,Rm)/Var(Rm) = ρσ jσ m / σ m2 = ρσ j / σ m


Cov(Rj ,Rm)
= 0.3*(0.15-0.135)*(0.20-0.116)+0.4*(0.09-0.135)*(0.05-0.116)+0.3*(0.18-.135)*(0.12-
0.116) = 0.00162
 Betaj = 0.00162/0.001485 = 1.09

σ XY
(d) Correlation = ρ = = 0.00162/[(0.0385)*(0.0622)] = 0.6765
σ XσY
Note: Betaj = 0.6765*0.0622/0.0385 = 1.09

2
2. Assume the single-index model holds. The variance of return on the market portfolio is
0.25. Stocks A and B have the following characteristics:

Stock Beta Residual variance


A 1.2 0.4
B 1.6 0.8
What is the variance of return on an equally weighted portfolio formed of stocks A and
B?

Solution: Var(rp) = Var(wA rA + wB rB) = wA2 Var(rA)+wB2 Var(rB)+2 wAwB σA,B


 Var(rp) = (1/2)2Var(rA) + (1/2)2Var(rB) + 2*(1/2)*(1/2)* σA,B

Var(Ri) = βi2 Var(Rm) + Var(ei) = βi2 σ2m + σ2


 Var(rA) = 1.2*1.2*0.25+0.4 = 0.76
 Var(rB) = 1.6*1.6*0.25+0.8 = 1.44

Cov(RA , RB) = Cov(αA+βARm+eA , αB+βBRm+eB)


= Cov(βARm , βBRm) = βA*βB*Var(Rm) = 1.2*1.6*0.25 = 0.48

Thus, Var(rp) = (1/2)2Var(rA) + (1/2)2Var(rB) + 2*(1/2)*(1/2)* σA,B


= 0.25*0.76+0.25*1.44+0.5*0.48 = 0.79

3
3. Assume the single-index model holds.
Asset Beta Covariance with the
market portfolio
A 1.2 0.3
B βB 0.5
(a) Find the value of βB.
(b) Suppose we form a new portfolio that has a 30% weight on asset A and a 70% weight
on asset B. What is the covariance between this new portfolio and the market
portfolio?
(c) Find the value of Cov(RA , RB).
(d) If the residual variance of asset A is 0.4 and the residual variance of asset B is 0.8,
what is the variance of return on the portfolio in part (b)?

cov(rA , rm ) 0 .3
Solution: (a) β A = 2
 1 .2 = 2
→ σ m2 = 0.25
σ m σ m

cov(rB , rm ) 0 .5
βB = 2
= = 2.0
σ m 0.25
(b) rp=0.3rA+0.7rB
Cov (rp, rm) = Cov(0.3rA+0.7 rB, rm)=0.3Cov(rA, rm)+0.7Cov(rB, rm)
= (0.3)*(0.3)+(0.7)(0.5) = 0.44

(c) Cov(RA,RB) = Cov(αA+βARm+eA , αB+βBRm+eB)


= Cov(βARm , βBRm) = βA*βB*Var(Rm) = 1.2*2*0.25 = 0.60

(d) Var(rp) = Var(wA rA + wB rB) = wA2 Var(rA)+wB2 Var(rB)+2 wAwB σA,B


 Var(rp) = 0.09Var(rA) + 0.49Var(rB) + 2*0.3*0.7* σA,B

Var(Ri) = βi2 Var(Rm) + Var(ei) = βi2 σ2m + σ2


 Var(rA) = 1.2*1.2*0.25+0.4 = 0.76
 Var(rB) = 2.0*2.0*0.25+0.8 = 1.80
 Var(rp) = 0.09Var(rA) + 0.49Var(rB) + 2*0.3*0.7* σA,B
= 0.09*0.76+0.49*1.80+0.42*0.60 = 1.20

4
4. Assume the single-index model holds. You own 200 shares of each of the following
two stocks as your portfolio.
Stock Standard deviation Beta Price per share
A 50% 1.6 $25
B 30% 0.8 $15
The standard deviation of return on the market portfolio is 25%.
(a) What is the residual variance of return on stock A?
(b) What is the variance of return on your portfolio?

Solution: (a) σ A2 = β A2σ m2 + σ 2 (e A )


σ 2 (e A ) = σ A2 − β A2σ m2 = 0.5 2 − 1.6 2 0.25 2 = 0.09

(200)(25) 25 5
(b) w A = = = = 0.625
(200)(25) + (200)(15) 40 8
3
 wB = = 0.375
8
Cov(rA, rB) = σ AB = β A β Bσ m2 = (1.6)(0.8)(0.252) = 0.08
 σ P2 = w A2 σ A2 + wB2 σ B2 + 2 w A wBσ AB
= ( 85 ) 2 0.5 2 + ( 83 ) 2 0.3 2 + 2( 85 )( 83 )(0.08)
=0.1478

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy