Seasonal exponential smoothing with damped trends

In this paper the authors apply the strategy of trend-damping to the popular Winters exponential smoothing systems for seasonal time series. Efficient model formulations are derived for both multiplicative and additive seasonal patterns. An algorithm is given to test the stability of the models in cases where predetermined smoothing parameters are used. Empirical results are presented to show that trend-damping improves ex ante forecast accuracy in seasonal data, especially at long leadtimes.