A lot of R is used in laying out the algorithms taught throughout the syllabus, but it’s not tested in the midterms or finals (you can make do with pseudocode). Still, having previous R exposure is handy for comprehending the algorithm design a bit better, especially since the take-home assignment would involve some coded simulation + analysis.
Content revolves around simulating samples from common distributions, simulating with Monte Carlo methods, reducing variance of sample estimates, and a little bit on psuedorandom number generation. Don’t think we actually touched on any stochastic optimization though. There were mainly just lecture sessions, except a few times when we covered tutorial questions in-class. The prof did show on-the-spot R demonstrations to better illustrate concepts/algos in ways beyond static notes.
The midterm and take-home assignment made up for the 40% CA, as in whichever score was better would be the CA score. The finals then made up for 60%. Both midterm + finals allowed one cheat sheet each. Personally I found these quite manageable, with the finals a bit trickier.
Head over to our Shop for more module content!
