EC3304: Econometrics II
Semester 2, AY2013/14
Time allowed: 2 hours
INSTRUCTIONS TO STUDENTS
1. This assessment exam contains SEVEN (7) questions and comprises THREE (3) printed pages.
2. Answer ALL questions.
3. Write the answers for each question on a new page.
4. This is a CLOSED BOOK examination.
5. Non-programmable calculators are allowed.
6. The total mark for this exam is 100.
EC3304
1. (10 marks) (True of False) Using a fixed-effects regression one cannot estimate the effect of an explanatory variable which is constant over time (e.g., gender). Provide a brief explanation.
2. (12 marks) One of your classmates estimates a probit model and tells you that she got a fraction correctly predicted of 0.7. She asks you what that precisely means.
What would you say to her?
3. (15 marks) Explain the difference between a sharp and a fuzzy regression discontinuity design.
4. (12 marks total) Answer the following questions regarding a random walk time series process and provide a brief explanation.
(a) (3 marks) Is a random walk process stationary?
(b) (3 marks) Is it weakly dependent?
(c) (3 marks) Is the first difference stationary?
(d) (3 marks) Is the first difference weakly dependent?
5. (9 marks total) Which of the following processes are stationary? Assume that
t
is
uncorrelated across time.
(a) (3 marks) Yt = 0.9Yt−1 − 0.2Yt−2 +
t
(b) (3 marks) Yt = 0.7Yt−1 + 0.2Yt−2 + 0.1Yt−3 +
(c) (3 marks) Yt = 0.7Yt−1 + 0.2t +
t
t
6. (12 marks total) Suppose that Yt follows the model
Yt = 1 + 2t + 0.5ut + 0.2ut−1 + 0.1ut−2 , where ut is a serially uncorrelated random variable with mean 0 and variance σu2 .
(a) (3 marks) Compute the mean and variance of Yt .
(b) (3 marks) Compute the first and second autocorrelations of Yt .
(c) (3 marks) Show that ρj = 0 for j > 2.
(d) (3 marks) Is this time series process stationary?
2
EC3304
7. (30 marks total) In the United States, injured workers receive a proportion of their weekly earnings