THE M2-COMPETITION - A REAL-TIME JUDGMENTALLY BASED FORECASTING STUDY

The purpose of the M2-Competition is to determine the post sample accu racy of various forecasting methods. It is an empirical study organize d in such a way as to avoid the major criticism of the M-Competition t hat forecasters in real situations can use additional information to i mprove the predictive accuracy of quantitative methods. Such informati on might involve inside knowledge (e.g. a machine breakdown, a forthco ming strike in a major competitor, a steep price increase, etc.), be r elated to the expected state of the industry or economy that might aff ect the product(s) involved, or be the outcome of a careful study of t he historical data and special care in procedure/methods employed whil e forecasting. The M2-Competition consisted of distributing 29 actual series (23 of these series came from four companies and six were of ma cro economic nature) to five forecasters. The data covered information including the September figures of the year involved. The objective w as to make monthly forecasts covering 15 months starting from October and including December of the next year. A year later the forecasters were provided with the new data as they had become available and the p rocess of predicting for 15 months ahead was repeated. In addition to being able to incorporate their predictions about the state of the eco nomy and that of the industry the participating forecasters could ask for any additional information they- wanted from the collaborating com panies. Although the forecasters had additional information about the series being predicted the results show few or no differences in post- sample forecasting accuracy when compared to those of the M-Competitio n or the earlier Makridakis and Hibon empirical study.