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, 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. Consider the following variation on the wages model from Assignment 2:

wAGESi  = F0  + F1AGEi  + F2AGE2i  + F3SCH00Li  + F4 FEMALEi

+F5 ETHBLACKi  + F6 ETHHISPi  + ei

i = 1, 2, … , 500

where the variables are defined as follows:

WAGES = current hourly wages in $

AGE = age in years

AGE2 = square of AGE

SCHOOL = years of schooling

FEMALE = gender of respondent (1 if female, 0 if male)

ETHBLACK = African-American (1 if African-American, 0 otherwise)

ETHHISP = Hispanic (1 if Hispanic, 0 otherwise)

together with the dataset A3Q1.dta, which is available in Brightspace.

a) Using STATA, obtain basic summary statistics for the variables listed above, and then copy and paste the output into your assignment. NOTE: You may need to generate some of these    variables from the variables supplied in the dataset.

b) Using STATA, estimate this model, and then copy and paste the output into your assignment.

c) Using STATA, compute the correlation matrix for the set of explanatory variables, and then copy and paste this correlation matrix into your assignment.

d) Using STATA to run the relevant regressions, calculate the VIFs for this model using the steps outlined on p. 234, together with the formula in equation (8. 16).

e) Using Klein’s Rule of Thumb (discussed in footnote 6 on p. 235), do any of the R2 values    from the auxiliary regressions run in part d) suggest the presence of multicollinearity? Explain.

f) Check your answers to part d), by using the STATA vif post-estimation command. Be sure to copy and paste the relevant output into your assignment.

g) Review your answers to parts b), c), d), e), and f), and then draw an overall conclusion as to whether there is any evidence of a multicollinearity issue in the model. Explain.

h) Can you suggest a remedy for the multicollinearity issue, if any, in this model? Explain.

B. Consider the following time-series demand function model for flowers, seeds, and potted plants:

lnFL0wt  = F0  + F1 lnDPIt  + F2 lnRPFL0wt  + et                                        t = 1, 2, … , 45

where the variables are defined as follows:

lnFLOW = natural log of consumer expenditure on flowers, seeds, and potted plants lnDPI = natural log of aggregate disposable personal income

lnRPFLOW = natural log of the relative price index for flowers, seeds, and potted plants and

RPFL0w =  () x100

where

PFLOW = nominal price index for flowers, seeds, and potted plants

PTPE = nominal price index for total personal expenditures

together with the dataset A3Q2.dta, which is available in Brightspace.

a) Using an appropriate set of STATA commands, estimate this log-log regression model, and then copy and paste the output into your assignment. Be sure to include ALL your STATA commands in this output. NOTE: At the beginning of your STATA program, be sure to tell STATA that this is a time-series regression and that “YEAR” is the time variable. (See p. 9- 1 in Chapter 9 of Using STATA: A Practical Guide.)

b) Using STATA, plot the residuals from your estimated regression equation in a line graph against YEAR. (To produce a line graph in STATA, you can just replace “scatter” with “line” in the graph command which I discussed in class.) Copy and paste your line graph into your assignment. Does your graph exhibit any signs of serial correlation? Explain.