Anyway, consolidation time. What has happened since my FYP? From an academic point of view, I'm glad that I took the challenge of attempting it. It made me realise that I do not like to do research at all. Reading papers, dissecting them into smaller parts (some people call it 'digesting' :p) and interpreting them aren't really my cup of tea. Still I pushed through with it and although the results were not as satisfying, I was glad I had the opportunity to try it. After all, life is about challenges and it is not every day that I can challenge something knowing that there is little to lose! :)
Well, that's enough about research. As most of you may know, I've embarked on the next stage of learning which is my Masters by coursework in Statistics. Why did I choose further studies? That is because I wanted to push myself to the limit that coursework can bring me to. (As stated earlier I didn't like research. ) Also, I wanted to see how much more Mathematics I could still learn at the higher level. After all, this is my 17th year of learning Mathematics. :) Now, let's introduce (drum-roll please) the modules that I've taken in my first semester at NUS:
- ST5214 - Advanced Probability Theory
- ST5210 - Multivariate Data Analysis
- ST5223 - Statistical Methods
- ST5206 - Generalized Linear Models
- EC5102 - Macroeconomic Theory
Instead, I will just describe my experiences studying for these modules.
For the first module, it was extremely hard-going. This was because the module concerned itself with a fair bit on real analysis which unfortunately, was not a required module in my undergraduate studies. It was an eye-opener for me as I started my journey into the depths of probability spaces. Imagine my astonishment when I realised that probability was not as simple as just calculating them. (Actually, the fact that my textbook read "A Second Course in Probability" by Sheldon Ross should have clued me in!) Sigma algebras, expectations and martingales were given new life. My sincerest thanks to my professor Adrian Roellin for making this course so interesting! The examinations was tough and I hope to do well.
The second module was more in line with what I have learnt before. Content-wise for the first half was manageable as it mainly dealt with multivariate normal distributions with a healthy dose of confidence intervals and hypothesis testing. The use of R was also applied extensively with a graded assignment depending on it exclusively. The mid-term was nit-picky if I may add, as it included information which wasn't emphasized on in class. No complaints as it was an open-book exam. The second half was more difficult as it was on Principal Components Analysis, Factor Analysis and Classification. These are three topics which I was greeting for the first time. Their application in R was also trying and I struggled to grasp them. Still, I'm glad I at least got to meet these three topics! The finals was suppose to be manageable although I made a mess of it. (First time I got a headache while doing an exam.)
The third module should be renamed Bayesian Models. Maybe it was because the professor's area of research was in Bayesian statistics that he decided to test us on that particular topic. At first glance, the notes was on linear models and was similar to regression analysis with generalized linear models thrown in. But his tutorial questions were a different story. (More likely, they belonged on Pluto rather than Earth.) Still, the students diligently went through his assignments and tutorials and was unexpectedly rewarded in the final exams as almost all the questions were similar, if not identical to them. There are gifts to be given for patience and practice! Lesson learnt: practice, practice and more practice will lead to some degree of proficiency.
Generalized linear models should have been a difficult module. After all the textbook used was around my age. The words used were archaic and reminded me of the papers I had to read for my FYP. As fate would have it, we were given a wonderful lecturer. Names shall not be revealed but he made it simple for which we were thankful. Many portions of the syllabus were omitted as they were deemed too difficult as it was a relatively new topic. Also, the final exams was changed from a close-book exam to an open-book one. Immensely grateful as a close-book exam would mean enormous amount of time cramming for it. Although we did not learn as much, the module was a great introduction on the applications of generalized linear models. I'm glad my friend suggested doing the past year papers as there was a question which was modified from there.
Finally, macroeconomic theory. Why did I choose it? Well, my course was designed to allow me to take 2 cross-faculty module and I chose economics as it had great synergy with mathematics. I initially wanted to choose the applied economics version of this module, but had to settle for the economics version as the applied economics version was changed to a self-funding course which meant I had to pay $2k plus to take that particular module. No regrets in taking that module as I had a great lecturer who really made the lessons come alive. His lessons always start with 20 mins of discussion about current economic topics that are relevant to what he is teaching. He asked us questions that really set us thinking like "Why did this table come from China?". He also set aside prizes (12 flash disks) at his own expense if we could answer a question that he posed every week. His mid-term was challenging in a mathematical way while his finals was designed to allow us to pass while differentiating the students with his second question. A module that I would definitely recommend if he is still teaching it.
That's all for my first semester at NUS. It's a rather long post and I hope you all have endured up to this point! :) I hope to blog more often since it has become more of a lost art rather than something which in vogue. ^ ^
