NUS module reviews AY21/22 Y3S1

NOTE: This was another fully online semester so reviews may vary from classes taken under physical settings.

This was my first semester taking on 18 MCs so study-wise, I thought it would be more chill so I did a part-time internship. Turns out it was quite heavy especially since the economics modules were very math-intensive.

Lecturer: Dr. JO In-hwan
Format of classes: 3 hours weekly LIVE (and recorded) lecture

  • Midterm (35)%
  • Problem set X1 (10%)
  • Data assignments x3 (5% x 3)
  • Group Project (30%)
  • Participation (10%)

This is basically a harder version of EC2102 and EC3102. EC4307 is mainly about the study of business cycles aka short-run fluctuations in the economy. The first two weeks were literally a recap of EC2102 and EC3102 so you can literally not pay attention and be fine. Afterwards, it covers the usual consumer and firm optimisation problem except that it now has uncertainty. Besides learning the two-period and infinite horizon models, we learn how to use recursive method to solve infinite horizon optimisation problems. Government optimisation problems are covered, but only a little. There is no IS-LM and AD-AS graph sketching in EC4307 at all. After these fundamentals are introduced, the lecture moves on to the measurement of business cycle and that’s when the data assignments come in. The last half of the module is spent learning Real Business Cycle theory, New Keynesian theory, DSGE model etc. Tbh it’s just the same as what we learnt in EC2102 except that now it’s math-centric.

Data assignments, which are done on Excel and Stata (only one), are really easy and you can discuss with your friends. Prof Jo does an extremely good job in guiding you and his instructions are very clear. These assignments are mainly on detrending a time series, extracting the cyclical component of a time series, summarising business cycle statistics and computing impulse responses. Problem set is a bit tough – think typical economics derivation question – but still doable.

Prof Jo’s midterm was closed-book, physical and it tests almost everything except for the last chapter since we had no time. The midterm had around 13 MCQs and 2 structured questions under 70 mins (iirc) so it’s quite intense. The midterm questions are mainly on derivation, few memory and really your knowledge of the classroom materials. Nothing too tricky and it’s doable for me. His past averages for midterms are a fail (legit not kidding. Our batch had an average of 17+ out of 40) so it really is ok if you find yourself struggling through the math.

Group project is just about analysing an economy’s macroeconomic issues, evaluating its current policies and then coming up with solutions. Report and presentation, nothing too difficult or different here. Tip is to pick a country where you can find data and information on.

A word of warning though. If you think EC3102 is tough and a big struggle, please DO NOT do this module because EC4307 is a big jump from EC3102. In a typical question, you can see a lot a lot of Greek letters and there are really no numbers. You are expected to know some linear algebra, time series concepts too although EC3304 is not a prerequisite. BUT I can reassure you that Prof Jo is a really really really good and nice prof. Easily the nicest prof I have met in NUS so far. During the midterm week, he had office hours till 8PM and to ensure students have fair chance to consult, each student only had (i think) 2 15-min slots to ask questions. I think it was a fair practice because some students have a lot to ask and it just isn’t fair imo. Prof Jo’s assignments and deadlines are not so demanding too. He is good at teaching too, deriving equations one-by-one and is always willing to help you understand the math. At our last lecture he even gave a 10-15 mins speech on life, on how we should deal with failure, try and be optimistic in this climate LMAO what a nice prof (better than Zhang Yang and Tim Wong!!)

Difficulty: 4.5/5 (A LOT OF MATH)
Workload: 2/5 (no weekly homework, no readings, no need class prep)
Recommendation: Do if nothing else to select. Otherwise content can be boring.

Lecturer: Dr. ZHONG Song Fa
Format of classes: 3 hours weekly LIVE (and recorded) lecture

  • Participation (5%)
  • Individual homework x 4 (10%) *Graded based on effort
  • Group research paper presentation (15%)
  • Group research proposal (15%)
  • Final (55%)

EC4394 is a module with many experiments and theories that debunk the usual classical economics. So instead of assuming people are selfish in game theory, behavioural economics also has these concepts like fairness equilibrium and inequity aversion model that explain why people behave fairly in lab and field experiments. There are a total of three big themes: Prospect theory (which goes against expected utility and Coase theorem), distributional preferences (which goes against standard Nash eqb predictions), and quasi-hyperbolic discounting (which goes against typical exponential discounting learnt in micro mods). Each theme takes around 2-4 weeks to go through and there are some homework to test your understanding.

Each week there is one compulsory reading and 3-5 supplementary readings, and each reading is around 10-20+ pages (it’s long because of the tables). Although the prof said it is not compulsory to read all, I strongly recommend to read because the lecture slides are literally based on this readings. It is almost impossible to self-study just by looking at the slides and listening to the prof so the intro and conclusion parts of each reading would greatly aid your understanding of what is going on in each experiment.

While the content is very very interesting, i find myself struggling to do the homework and finals because they were too applied. Unlike the usual economics modules, EC4394 requires a lot of logic and understanding. The questions usually are about an experiment and you are expected to either predict or explain the outcomes in a behavioural manner. Then you need to use certain concepts to prove whether that model can predict that outcome or not. It was tough for me because the answers can be very broad and subjective; there isn’t really a fixed answer and I felt it was tough for me especially since there isn’t anyone to mark your homework.

The group project was horrible since my groupmates were not the best. The first one was just a tutorial presentation on a reading and pretty much no one ever asked a Q&A so yeah. It was quite rushed because you only have one week to prepare since the reading you present is linked to the topic taught that day. The second project was a research proposal and you need to come up with a research proposal related to any behavioural concepts. It can be an easy and fun module if done with the right team. Otherwise, the assignments for ec4394 is very light.

Prof Zhong is a nice prof but imo, I feel his teaching can be better especially when he goes through the math concepts and the homework answers. Sometimes we just don’t get how he derived an inequality and there were some typos in his answer keys. Funny part was some questions had N parts but his answer key only had N-1 answers ??? Nonetheless he is very free and open to consultation and email clarifications so that’s good.

Difficulty: 4.5/5
Workload: 3.5/5 (4 homework in total, weekly readings)
Recommendation: Do if you are very well-versed in EC3101, independent and have a lot of time since there are quite a few readings.

Lecturer: Dr. Sorawoot Srisuma
Format of classes: 2 hours weekly LIVE (and recorded) lecture, bi-weekly 2 hours tutorial

  • Participation (20%)
  • Midterm (20%)
  • Final (60%)

Really dislike this module. It is another compulsory core module and this time round, it covers way more topics than Kelvin did in 3303. EC3304 is on panel data, fixed effects regression model, binary choice models, instrumental variable regression models, and time series. You learn a different set of LSA assumptions (yes this again) for each theme and you learn the proof/intuition behind each theme and the Stata commands. FYI EC3304 under this prof is quite Stata-heavy and every tutorial has at least one Stata question so it’s good for those who hates proving.

For the first few weeks, you learn panel data and fixed effects regression models. Then it is about binary choice model (probit/logit/NLLS models). Post-midterm material is on instrumental variables, 2SLS. The highlight for me was the time series theme which specifically focuses on AR(P) and ADL(P,Q) models (NO ARIMA though). I find instrumental variables the toughest to understand simply because it involved a lot of hypothesis tests, reduced form equations and structural equations. It’s only in this module did I realise that OLS is just one estimation technique; you learn things like GLS and MLE here because OLS and simple linear regression are simply too simple LOL.

I particularly did not enjoyed the lecture at all and I found it a huge struggle trying to understand what the prof was trying to say. More than often I find myself struggling to filter out those filler words (“erm, yeah, right” etc) just to connect the meaning of the sentences. I stopped going for live lectures after a while especially when I know it’s going to be a concept-heavy lecture. Sometimes it’s so bad i don’t understand whether the prof is talking or teaching; emailing or approaching TA are better options since there is no accent barriers. Bear in mind the prof teaches some concepts from SW textbook but does not tell you where he plucks the info from so it’s just??? There isn’t a recommended reading list so you have to be very independent and do a lot of Googling.

In terms of content wise, I would say that EC3304 is very difficult and much more difficult to understand than EC3303. Else you can always consult him but for our batch, the consultation slots weren’t limited so you can find yourself waiting for almost an hour (not joking here) for students who came before you to clarify finish their questions. Pretty bad initiative and def not considerate imo. Nonetheless, once you understand the concepts in EC3304 really isn’t anything else you need to revise for finals because the concepts are just so little. Thankfully this module does not test on proofs at all, unlike EC3303, so you need not memorise anything. Also, Dr Sorawoot’s questions are really very straightforward and you just need to substitute in numbers into models, apply the right formulae etc for the open-book midterms and finals.

Difficulty: 4/5
Workload: 1/5 (bi-weekly tutorials. Lots of self-revision required)
Recommendation: no choice for economics majors. Please revise hypothesis tests.

Lecturer: Dr. WANG Senhu, Dr. MU Zheng
Format of classes: 2 hours weekly LIVE (and recorded) lecture, bi-weekly 2 hours tutorial

  • Literature review (25%)
  • Qualitative interview and report (30%)
  • PSPP quantitative report (30%)
  • Tutorial participation (15%)

While an open module for all undergraduates, SC2101 is very sociology imo and there were many times during the lectures and assignments where I struggle to do well in because I just didn’t understand how to write a sociology paper. The lectures were quite boring and dry because it’s just about concepts ON research; didn’t gain much out of it and they weren’t really useful for the assignments. But i would say the tutorials are very helpful since you are literally doing your assignments during the tutorial sessions.

In terms of assignments, while I felt it was doable I think it was tough especially for the qualitative one. The qualitative project requires you to focus on a topic, interview two people on how COVID-19 impacted their decision, then write a report on it. It was tough because some students may not have a representative or even valid people to interview from and this adds another problem. also, the word count is another headache since we had to include both verbatim AND the actual report. Didn’t made sense to me and despite my best attempts to cut, my TA said that more can be elaborated?? Same goes for the analysing of the lit review of two research papers. The word count is simply too little for a good analysis.

The quantitative assignment requires you to download PSPP, which is essentially a free version of SPSS, and to analyse some socioeconomic variables in a given dataset. Once again, the questions are so vague you don’t really know what the TA wants.

Good thing about this mod is that it teaches some research skills especially for those who have zero research skills. I took this module to clear UE and hoping to deepen my skills but apart from the qualitative coding experience, nothing much was taken away from this experience.

Difficulty: 3/5 (need to think)
3/5 (need to do a few prep work before tutorials)
Recommendation: Not if you are looking for a light-workload UE mod.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s