ECOM30001: Basic Econometrics

2 × Group Assignments, 10% each
1 × Mid Semester 40 minute Multiple Choice Exam, 10%
Tutorial Participation, 10% (5% Attendance, 5% attempting pre-tute work)
2 hour End of Semester Exam, 60% (Hurdle Requirement)

NB: Lecture/tutorial content and assignments/exams are identical to the masters subject with the same name (ECOM90001), however masters students don't get tutorial participation marks and instead their exam is worth 70%. This subject and ECOM20001/Econometrics 1 are also a non-allowed subject pair.

Can be done in groups of 1-4 with people in your tutorial. Despite the horror stories you hear about group assignment, I honestly found doing it in a group helped with certain things I didn't understand yet, although everyone contributed fairly which isn't guaranteed to happen (But if your first group isn't good you don't have to work with them on the second assignment). Assignments were a bit more challenging than the exam IMO (although this may have been from my understanding during the semester vs during the final exam), and as mentioned above required you/your group to create your own R script, which must be added as an appendix to the assignment. That being said, marks were pretty good overall (From LMS, assignment 2 average was 30.73076/35, median was 31.00/35).

Overall I found this to be an interesting and well-organised subject. Some of the content was fairly difficult to grasp, but the assessment itself was not too bad at all, and mainly required knowledge/understanding of the various properties and procedures taught, and well as interpretations. It was often stressed in lectures that we don't need to reproduce the complex proofs and algebra presented, and that it was just there so we could see where things came from. R is the language used, which is a plus as this is one of the languages used more in the "real world". Downloading R studio (it's free ) is very recommended, if not vital. However, you're not expected to come up with your own code or anything, you'll pretty much just need to copy certain aspects from the lecture slides and/or tutes for the assignments (which can done in groups, but I'll assume no one who's taking the time to read subject recommendations wants to be "that person" who doesn't contribute ).

Major topics were:
-The Basic Linear Model (Statistical properties, hypothesis tests, model specification)
-Dummy (Indicator) variables
-Heteroskedasticity
-Autocorrelation
-Time Series
-Stochastic Regressors (Issue of Cov(X,ε) ≠ 0)
-Panel Data
-Count Data
-Binary Outcomes

Focuses on the major topics listed above from Dummy Variables onwards. Apart from undertaking F- and t- hypothesis tests, the content from the mid semester exam is not directly examinable. As was the case with the mid sem, the practice questions provided were much more difficult and theoretical than the actual exam. The 2018 exam was a fair representation of what to expect. However, as stated on the LMS, exams solutions are intentionally far more extensive than what is needed for a "perfect answer".
No real surprises on my exam. The focus is on understanding/explaining properties of the models/variables, the implications of violating the standard MR assumptions, basic calculations/interpretations and tests. R output is provided to answer questions, but everything you need from it is very straightforward (especially since you've seen it all semester). There is also a formula sheet (for the mid sem too), meaning you don't have to remember the exact detailed of formulae/models, although you still need to know how to use them.

Yes, with screen capture, although I think in a lecture theatre with dual screens you may miss out on some of the stuff currently being projected. You'll have access to everything, but this may make the actual lecture a little less clear.

Andrew (Andy) Clarke

Involved being talked through the slides/theory, with examples that often included R output. Extra accompanying handwritten notes were also often used and posted to the LMS. Andrew was a great lecturer and explained things well, although (as is the case with most uni subjects) extra time is needed to review, process and understand everything. There was quite a lot to sort through in the lecture slides which made reviewing it a bit harder, but it wasn't a major issue.

I expected this to be online since it was MC and so many other economics subjects do this, but it was treated as a formal mid sem in Wilson Hall under exam conditions. Focused on the 'Basic Linear Model' topics. Practice MC questions were provided, which were much harder than the actual test. 40 minutes was more than enough time to complete and check everything. I believe Andrew said the median was 10/12 and the mean was close to that too.

2018 final exam and practice questions for the final and mid sem exam provided, all with extensive solutions. 2016 past exam also available on library website, but no solutions.

4 Out of 5

No mandatory requirements, but lectures reference Principles of Econometrics, 5th ed by Hill, Griffiths, and Lim. I was able to find the 4th ed online, and barely used it. A few others are also recommended in the subject guide, but it's all optional and buying any textbook is definitely not necessary.

Very R focused. You're given access to the questions and R script beforehand and expected to use the script to

2 × 1 Hour Lecture per week, 1 × 1 Hour Tutorial per week

Semester 1, 2019

H1 (85)