Where does knowledge about forecasting come from?
The short answer to this question is research. Research on forecasting has produced many changes in recommended practice, especially since the 1960s. Much advice that was formerly given about the best way to generate forecasts has been found to be wrong. For example, the advice to base forecasts on regression models that fit historical time-series data has had a detrimental effect on accuracy. Sometimes the research findings have been upsetting to academics, such as the discovery that relatively simple models are more accurate than complex ones in many situations. Perhaps the major reason that research has been so important in forecasting is that it has stressed empirical results that compare the forecasting performance of alternative methods. One of the more important empirical comparisons was the M-competition (Makridakis, et al 1982.). The M-competition was followed by others, the most recent being the M3-Competition (Ord, Hibon, and Makridakis 2000). Emphasizing empirical findings
