search query: @author McKenzie, E. / total: 8
reference: 3 / 8
| Author: | Gardner, E. S. Jr. McKenzie, E. |
| Title: | Model identification in exponential smoothing. |
| Journal: | Journal of the Operational Research Society
1988 : SEP, VOL. 39:9, p. 863-867 |
| Index terms: | FORECASTING TIME SERIES MODELS |
| Language: | eng |
| Abstract: | In practice, model identification is traditionally ignored in forecasting via exponential smoothing. It is common to apply the same exponential smoothing model to every time-series in a collection. The aim is to develop a robust procedure for identifying exponential smoothing models. The procedure should also be simple and efficient so that it can be automated in large forecasting applications such as in inventory control. The procedure is based on a comparison of the variances of relevant differences of the data of time-series. The order of differencing yielding maximum variance suggests an appropriate model for the time-series. |
SCIMA