Submitted

DONE! At 12.01am.

Deadline is 12 noon. But I can't bear keeping it with me any longer. I need to get rid of it and I did. After working on this project for almost a month, I am finally relieved from the torture chamber. I am amazed how some people can just spend months looking at data set, generating results. Takes a lot of patience. Unfortunately, I am not those kind. I'll rather manipulate my data if I can. Oops.

It's pretty crazy. I started with a simple analysis on the relationship between spot and futures prices in the crude oil market. I picked this title based on my naive belief that since the data is easy to obtain, I will have no problem working on it. Little did I know that it brings me to a whole lot of dead-ends till I have to abandon the project two weeks before due date to start on another topic which is to correct for the problem I found in the first place. Apparently, the non-normality in my residuals are due to the presence of volatility clustering in the series itself (which is proven by a number of findings). Therefore, correcting it is beyond my ability and perhaps the scope of the course.

That was a big risk since the lecturer had just started teaching the methodology. So, I had to do a lot of readings on my own. I just couldn't bear going back to attempting to achieve success in obtaining a correctly specified VAR. It was not a decision to be regretted. ARCH and GARCH is not that difficult after all since I am not trying to re-model the whole regression. Just too bad my lecturer never did went into much details. So, I had a bit of hard time trying to decipher the results.

Nevertheless, it has reached an end...for now as I fear my next two projects could be related to testing PPP or UIP. Sigh...