Why not subscribe?

Thursday, August 19, 2010

Forecasting contests

Think you can forecast?

There have been forecasting competitions before (notably the M1, M2 and M3 contests organized by Makridakis and colleagues).  There was also the well-publicized Netflix competition.  Now there's a site devoted to these.

http://kaggle.com/ organizes the data sets and keeps track of the results. Some current projects involve trying to improve on the Elo system (used to rate chess players and predict the outcome of chess games) and a set of tourism time series organized by Rob Hyndman, the editor of the International Journal of Forecasting.

The major conclusion from the M1, M2 and M3 competitions -- that relatively simple models perform best, if you have relatively simple data (univariate time series) -- is a great one for practitioners and I have appreciated the published empirical support.

A secondary aspect is to provide standardized comparison sets of data for evaluating new forecasting schemes. We know how many methods perform on hundreds of data sets, so new methods can be compared on a simple and honest basis.