A Bayesian multiple models combination method for time series prediction

In this paper we present the Bayesian Combined Predictor (BCP), a probabili stically motivated predictor for time series prediction. BCP utilizes local predictors of several types (e.g., linear predictors, artificial neural ne twork predictors, polynomial predictors etc.) and produces a final predicti on which is a weighted combination of the local predictions; the weights ca n be interpreted as Bayesian posterior probabilities and are computed onlin e. Two examples of the method are given, based on real world data: (a) shor t term load forecasting for the Greek Public Power Corporation dispatching center of the island of Crete, and (b) prediction of sugar beet yield based on data collected from the Greek Sugar Industry. In both cases, the BCP ou tperforms conventional predictors.