论文信息 - Computing the variance of the forecast error for the holt‐winters seasonal models

Computing the variance of the forecast error for the holt‐winters seasonal models

This paper presents expressions for the variance of the forecast error for arbitrary lead times for both the additive and multiplicative Holt-Winters seasonal forecasting models. It is shown that even when the smoothing constants are chosen to have values between zero and one, when the period is greater than four, the variance may not be finite for some values of the smoothing constants. In addition, the regions where the variance becomes infinite are almost the same for both models. These results are of importance for practitioners, who may choose values for the smoothing constants arbitrarily, or by searching on the unit cube for values which minimize the sum of the squared errors when fitting the model to a data set. It is also shown that the variance of the forecast error for the multiplicative model is nonstationary and periodic.

Arnold L. Sweet | A. Sweet

[1] Ed McKenzie. A Comparison of Some Standard Seasonal Forecasting Systems , 1976 .

[2] Arnold L. Sweet. Adaptive Smoothing for Forecasting Seasonal Series , 1981 .

[3] John O. McClain. Dynamics of Exponential Smoothing with Trend and Seasonal Terms , 1974 .

[4] Douglas C. Montgomery,et al. Forecasting and time series analysis , 1976 .

[5] J. Brenner,et al. Difference Equations in Forecasting Formulas , 1968 .

[6] L. Debnath,et al. Transform methods with applications to engineering and operations research , 1980, Proceedings of the IEEE.

[7] John O. McClain,et al. Response-Variance Tradeoffs in Adaptive Forecasting , 1973, Oper. Res..

[8] Peter R. Winters,et al. Forecasting Sales by Exponentially Weighted Moving Averages , 1960 .

[9] H. Theil,et al. Some Observations on Adaptive Forecasting , 1964 .

[10] Ed. McKenzie. The Monitoring of Exponentially Weighted Forecasts , 1978 .