MODEL SELECTION IN UNIVARIATE TIME-SERIES FORECASTING USING DISCRIMINANT-ANALYSIS

When a large number of time series are to be forecast on a regular bas is, as in large scale inventory management or production control, the appropriate choice of a forecast model is important as it has the pote ntial for large cost savings through improved accuracy. A possible sol ution to this problem is to select one best forecast model for all the series in the dataset. Alternatively one may develop a rule that will select the best model for each series. Fildes (1989) calls the former an aggregate selection rule and the latter an individual selection ru le. In this paper we develop an individual selection rule using discri minant analysis and compare its performance to aggregate selection for the quarterly series of the M-Competition data. A number of forecast accuracy measures are used for the evaluation and confidence intervals for them are constructed using bootstrapping. The results indicate th at the individual selection rule based on discriminant scores is more accurate, and sometimes significantly so, than any aggregate selection method. (C) 1997 Elsevier Science B.V.