Technical Note—Renormalization of Seasonals in Winters' Forecasting Systems: Is it Necessary?

Normalizing the initial estimates of the seasonal factors in Winters' seasonal forecasting systems is a universally recommended practice. Once forecasting begins, however, the seasonal factors lose their normalization. It is often recommended that these factors be renormalized either seasonally or continuously, i.e., with every observation. We show that this latter procedure is never necessary in the case of the additive model, since a simple change in the smoothing constants can achieve the same ends. We also give a simpler and more efficient method of obtaining seasonally renormalized forecasts for this model.