You are on page 1of 3

Name:

FIN 235-A Advanced Statistical Modeling in Finance


Fall Semester, 2023
Midterm Exam

1. (40 points) The single-index model (SIM) is possibly the simplest asset pricing model
that we use in finance to measure the tradeoff between the systematic risk and return
of a stock. Mathematically the SIM could be expressed as

rit − rf = αi + βi (rmt − rf ) + ϵit , (1)

where rit is the (simple) return for security i at time t, rmt is the market return at
time t, rf is the risk-free rate, and ϵit is the error term. The coefficient βi measures
the systematic risk that firm i is facing.

(a) Using the data in “Data ForFIN235MidtermExam.xlsx”, please estimate αi and


βi via OLS regression in Excel. (Note: Please crop the Excel regression output
and paste it below.)

(b) Explain the estimated βi in words.

1
2. (30 points) Suppose that you had estimated the following market model on an asset:

rt = α + βrmt + ϵt , (2)

where rt is the (simple) return at time t, rmt is the return on a proxy for the market
portfolio at time t, and ϵt is the error term. The coefficient β measures the systematic
risk of this asset. You found that the estimated beta was 0.89 with a standard error
0.33.
A security analyst has told you that this asset has a return that is riskier than
the market portfolio. The model is estimated over 62 daily observations.

(a) Write out the null and alternative hypotheses of the test.

(b) Test this hypothesis at the 5% level. (Hint: t5%,60 = 1.671; t2.5%,60 = 2.000.)

(c) What is your conclusion?

2
3. (30 points) Suppose a researcher wants to test stock returns (y) on four factors (x2 ,
x3 , x4 , and x5 ). The regression equation is

yt = β1 + β2 x2t + β3 x3t + β4 x4t + β5 x5t + ut .

The regression is carried out on 205 monthly observations. The researcher believes
that the sum of β2 and β3 equals 1, and β4 equals 2.

(a) Write out the null and alternative hypotheses.

(b) Write out the restricted and unrestricted regression equations.

(c) If the two RSS are 340 and 480, test the null hypothesis at 5% level.
Note: F(2,200) = 3.00 (5% level) and 2.30 (10% level).

You might also like