Scribd
Upload
96 views3 pages

2016 Midterm 1

This document provides instructions and questions for a financial econometrics midterm exam. It includes 4 exercises assessing skills in: 1. Interpreting regression output and hypotheses testing based on t-ratios from a cross-sectional stock return regression. 2. Identifying potential concerns with variable selection and suggesting alternative procedures. 3. Comparing regression metrics like RSS, R-squared, and adjusted R-squared between levels and changes specifications with a lagged dependent variable. 4. Discussing appropriate model specifications and tests for panel data examining the effect of manager experience on mutual fund returns, including potential endogeneity issues.

Uploaded by

dro land
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
96 views3 pages

2016 Midterm 1

This document provides instructions and questions for a financial econometrics midterm exam. It includes 4 exercises assessing skills in: 1. Interpreting regression output and hypotheses testing based on t-ratios from a cross-sectional stock return regression. 2. Identifying potential concerns with variable selection and suggesting alternative procedures. 3. Comparing regression metrics like RSS, R-squared, and adjusted R-squared between levels and changes specifications with a lagged dependent variable. 4. Discussing appropriate model specifications and tests for panel data examining the effect of manager experience on mutual fund returns, including potential endogeneity issues.

Uploaded by

dro land
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 3

ECON 818: Financial Econometrics

Midterm #1
February 2nd, 2016

You need to justify your answers carefully and show your work in order to get full credit.

Partial credit may be given for each question.

1
Exercise 1: [Brooks 6p177]
You estimate a regression of the form below in order to evaluate the eect of various rm-
specic factors on the returns of a sample of rms:

ri = β0 + β1 Si + β2 M Bi + β3 P Ei + β4 BET Ai + ui

where
• ri is the percentage annual return for the stock,

• Si is the size of the rm i measured in terms of sales revenue,

• M Bi is the market to book ratio of the rm,

• P Ei is the price/earnings (P/E) ratio of the rm,

• BET Ai is the stock's CAPM beta coecient.

You run a cross-sectional regression with 200 rms, and you obtain the following results
(with standard errors in parentheses):

r̂i = 0.080 + 0.801Si + 0.321M Bi + 0.164P Ei - 0.084BET Ai


(0.064) (0.147) (0.136) (0.420) 0.120

1. Calculate the t-ratios. What do you conclude about the eect of each variable on
the returns of the security? On the basis on your results, what variables would you
consider deleting from the regression?
2. If a stock's beta increased from 1 to 1.2, what would be the expected eect on the
stock's return? Is it the sign on beta you would be expected?

Exercise 2: [Follow-up exercise]

1. Do you have any concerns regarding the selection of variables suggested in the previous
exercise? Be specic and justify your answer.
2. Can you suggest an alternative procedure that would mitigate some of your concerns?

2
Exercise 3: [Brooks 7p177]
You estimate the following econometric models including a lagged dependent variable,
yt = β1 + β2 x2t + β3 x3t + β4 yt−1 + ut
∆yt = γ1 + γ2 x2t + γ3 x3t + γ4 yt−1 + vt

where ut and vt are iid disturbances.


1. Will these models have the same values for
(a) the residual sum of squares RSS;
(b) the R2 ;
(c) the adjusted R2 .
Explain carefully your answers in each case.

Exercise 4:
Imagine you have been given the task of evaluating the relationship between the return on a
mutual fund and the number of years its manager has been a professional. You have a panel
data set which covers all of the mutual funds returns in the years 1970-2010. Consider the
regression,
ri,t = α + βexperi,t + ϵi,t
where ri,t is the return on fund i in year t and experi,t is the number of years the fund
manager has held her job in year t.
The initial estimates of α and β are computed by stacking all of the observations into a
vector and running a single OLS regression (across all funds and all time periods).
1. What test statistic would you use to determine whether experience has a positive eect?
2. Suppose that experience has no eect on returns but that unlucky managers get red
and thus do not gain experience. Is this a problem for the above test? If so, can you
comment on its likely eect?
3. Could the estimated β̂ ever be valid if mutual funds had dierent risk exposures? If
so,why? If not, why not?
4. If mutual funds do have dierent risk exposures, could you write down a model which
may be better suited to testing the eect of managerial experience than the initial
simple specication? If it makes your life easier, you can assume there are only 2
mutual funds and 1 risk factor to control for.

You might also like