Hi, just as a general note I'm slowly incrementally adding code towards doing linear/bilinear regression [here](https://github.com/davidtweed/multicoreBilinearRegression), but it's very slow going (primarily because I'm only able to spend about 30 min per day on it while on the train, which isn't really long enough to fully engage my "concentration mode". Still not at the point of being a working program. I still hope to get _something_ finished before I run out of time.
In terms of things to predict, the only real thoughts I've had so far is that I'm reluctant to try to predict a 3-month average based El Nino 3.4, purely because it's likely to be noisy and hence errors aren't necessarily indicative of errors on the bigger problem. I'm inclined to try something like the (1+5+1)=7 month average El Nino 3.4 index for the following period immediately after the observatoins used for prediction, but that's not really more than a rough guess.
A quick note about prediction errors: it may be worth considering things other than squared-error as predictors. As I drone on about in the blog thing I'm working on, squared loss "wants to" reduce any big error to be smaller, even if that means inflating all the other errors up to that level. Sometimes an absolute-value-of-errors, or other alternatives, can be more insightful about classifier performance.