Assignment #1
ECON 323: Econometric Analysis 2 – Winter 2020
Due January 23, 2020, 10am in class
Instructions: While cooperating on the assignment is encouraged, pla- giarism is not. I will only accept hand written assignment submitted in person. DO NOT SUBMIT YOUR ASSIGNMENT ELECTRONICALLY. No late assignment will be accepted; late assignments (after 8:30am on the due date) will receive a mark of zero. Show your work as no marks will be allocated for the final answer alone.
Use Stata or R to do these. If you choose to use another software, please get my approval at least a week before the assignment is due.
Question 1
Using the Retirement Survey 1975 available from Odesi, answer the fol- lowing questions.
(a) Regress whether the indivdidual receives a pension from their em- ployer (consider this to be a continuous and normally distributed variable for the purposes of this question at this stage) on marital status, the province of residence, gender and the age the individual was when they left their longest employer . Report your results in equation form.
(b) Together, how much of the variation in marital status and procvince of residence explain?
(c) If an observation in the sample is such that the individual is a married female that lives in quebec and stayed with their longest ever employer until the age of 65, what is the predicted probability that they receive a pension from an employer?
(d) This individual receives a pension. Calculate the residual.
(e) Perform a t-test of the equality of the coefficient of age to 0.01 at 5% of statistical significance.
(f) Should you include gender and age in your regression? Discuss. (g) Is heteroskedasticity present in this model?
1
(h) Now treat whether individuals have a pension from their employer as the type of variable it should be and go back to assuming homoskedasticity, regardless of your results in (g). Repeat questions (a), (e), and (f) using the correct estimation method and report the ¡°new answers¡±. Has anything changed? Why?
(i) What is the average marginal effect for age at which they left their longest employer in this regression? Why should you consider the AME rather than the coefficient in this situation?
2