20.5 teaching hours cramped into four days - that's crazy coz I don't have much time for a breather. I think by Friday, I'll be totally exhausted. Especially since the 20.5 hours does not include preparation for lectures and tutorials, marking of assignments, preparing the questions for the exams and tests, extra classes for weaker students...gee I think I'll be living in SP for the next couple of months.

Okay, that's my ranting for the coming school term which starts next week. So, I don't know if I can update as frequently as before. Hopefully over the weekends.

Meanwhile, my colleague and I managed to get our Linear Programming model for Portfolio Optimization of Mutual funds working. The model is based on maximizing returns obtained through forecasting using moving averages. We tested it for 3 funds that Prudential Singapore offered. Depending on our risk tolerance, we managed to get a profit as compared to just savings alone. Also the profit is at least as comparable or even exceeds those obtained by just investing in one type of funds that appeal to our risk tolerance. Of course, now, we've got to do some validation and confidence analysis. We are still think of how to incorporate the standard deviation as a measure of risk in the constraints. If anyone has some ideas, I wouldn't mind hearing about it.

It is a simple LP problem with iterative updates as we near the horizon we've set. While it may not compete with the models used by experts, it is understandable enough for laymen investors; the ones that we're targeting the research at...after all the two of us are basically laypersons ourselves in this field of investments.

We hope that it is robust enough to test on the full suite of funds provided by Prudential and DBS and we're still studying the various parameters and variables that influence the model. Anyone like to try our model out later?

Cheers,

Sid