Exam Notes

1. I will return grades and details about projects asap. Projects received so far look really good. I am glad that people took them seriously and this will benefit you in the rest of your studies particularly your thesis

2. Apologies for confusion on last assignment. I got the assignment printed too late and you guys had got the impression it was cancelled. As always, I hope the common sense solution to this is ok with people. I will give full weight to the first two assignments i.e. these now count as 5 per cent each rather than 3.3 per cent each. If you really want to do the third assignment then simply individually email me and I will arrange a time.

3. People have been emailing for "exam hints". I was pretty clear at the lecture. There will be no trick questions in the exam. This is a long and detailed course but the exam is quite structured so if you have been following the course and assignments you will be more or less fine. Below is pretty much all I can say and while there is a lot of work to do for the exam I dont think I can be possibly clearer on the tasks and I have even provided the links to all the past and sample exams on the side bar to this blog. A video that may assist you in preparing for the exam is linked below. Good luck!!

Link to Useful Video

a. There will be five questions.

b. The first question is on regression output covering ols as well as heteroscedasticity and autocorrelation. It is mandatory

c. you then answer 2 from 4. The topics are maximum likelihood, instrument variables, cldv's and time series

d. time series will involve questions on autocorrelation testing, var, stationary mulitivariate regression models, cointegration testing.

e. Past exam papers are a good guide but by no means are questions guaranteed to repeat

f. Verbeek is the main source of examinable material as well as my lecture notes. The returns to education and VAR articles are also assigned readings.