Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

ECON 2560 – Applied Econometrics – Spring 2023

Homework #5

Due date:  Friday Feb 24, by 11:59 pm

Submission of late work:  If you miss the due date, you may submit your assignment within the next 24 hours, although there is a penalty for late submission.

· You may complete this assignment individually or within a group; in either case every student must turn in their own copy.

o Please upload your assignment answers into Canvas.  Your answers can be in Word, PDF, or Excel.

· Show work for partial credit.

1.  Using the data in salaryetc.xlsx in Canvas/Files,

a) Use OLS to estimate a regression of ln(salary) against educ, age, and sex and report the key results in matlab.

b) Test for heteroskedasticity using the alternative form of the White test (Wooldridge equation 8.20, p. 272, regressors yhat and yhat^2): what is the value of the test statistic?

c) What distribution does this test statistic have under the null hypothesis?

d) What is the p-value?

e) Create scatterplots of education vs squared residuals, age vs squared residuals, and sex vs. squared residuals. What insight do they provide into the nature of the potential heteroskedasticity?

f) Estimate the equation again using matlab's hac function to get heteroscedasticity-robust standard errors, and report the coefficients (should be the same as in a) and standard errors.

 2.  Using the regression you estimated in question 1, perform a RESET test.

a) What is the null hypothesis in this test?

b) What is the value of the test statistic?

c) under the null hypothesis and standard assumptions, what distribution does this test statistic have?

d) Do you reject the null hypothesis at a 5% significance level?