Simon Power

ECON 2220 E Fall 2021 Simon Power
Assignment 2: Due November 10
PLEASE BE SURE TO READ THE DOCUMENT ENTITLED “GENERAL ASSIGNMENT
GUIDELINES” BEFORE YOU BEGIN THIS ASSIGNMENT. ALL REFERENCES ARE TO
THE 7th EDITION OF STUDENMUND. UNLESS SPECIFIED OTHERWISE, PLEASE USE A
5% SIGNIFICANCE LEVEL FOR ALL TESTS. ASSIGNMENTS SHOULD BE SUBMITTED
THROUGH BRIGHTSPACE EITHER ON OR BEFORE THE DUE DATE.
IF IN DOUBT, PROVIDE MORE DETAIL IN YOUR ANSWERS, RATHER THAN LESS.
NOTE: If a statistical table does not give an entry for the appropriate number(s) of degrees of
freedom, then use the closest number(s) of degrees of freedom available.
A. Answer Question 3 of Chapter 5 (p. 149).
B. Answer Question 5 of Chapter 5 (p. 149).
C. Consider Question 7 of Chapter 3 (pp. 87-89).
Answer the following questions:
i) Obtain basic summary statistics for the IPOD3 dataset variables PRICE, NEW, SCRATCH,
and BIDRS using the STATA “summarize” command, and then copy and paste the output into
your assignment.
ii) Re-estimate the model using STATA, and then copy and paste the output into your
assignment.
iii) Report your regression results in the standard format of equation (3.2) on p. 72. (Use two
decimal places for the estimated coefficients, the standard errors, and the t-statistics.)
iv) Carefully explain the real-world meaning of the three estimated slope coefficients.
v) What signs would you have expected for each of the estimated slope coefficients? Carefully
explain your reasoning in each case.
vi) Based on your answer to part v), use t-tests to test these expectations about the signs of the
estimated slope coefficients. Be sure to write down the null and alternative hypotheses in each
case, and to explain whether you should reject or not reject the null hypothesis in each case.
vii) Test the overall significance of the regression using the step-by-step procedure outlined in
the example on p.145. Be sure to write down the appropriate null and alternative hypotheses, and
to explain whether you should reject or not reject the null hypothesis.
2
viii) Repeat part vii), but this time using the (more) general F-test formula given in equation
(5.10). Be sure to write down the appropriate null and alternative hypotheses, and to explain
whether you should reject or not reject the null hypothesis.
ix) Using the standard formula in equation (5.9), construct 90% 2-sided confidence intervals for
each of the estimated slope coefficients. Be sure to show your working, step-by-step.
x) Using an appropriate STATA command(s), check your answers to part ix). Copy and paste the
relevant STATA output into your assignment.
xi) Re-estimate the model using STATA, but this time omit the SCRATCH variable, and then
copy and paste the output into your assignment.
xii) Report the regression results from part xi) in the standard format of equation (3.2) on p. 72.
(Use two decimal places for the estimated coefficients, the standard errors, and the t-statistics.)
xiii) Using the discussion of four important specification criteria on p. 166-167, compare the
original model with a more parsimonious model which excludes SCRATCH. Which model do
you think is better? Carefully use the four specification criteria to make your decision, being sure
to state which criteria support your decision and why.