Stochastic models for forecasting (Q580862)

In this study the simple moving averages, exponential, and mixed exponential-simple moving averages models are explored in a forecast setting. In the development of the models, the statistical properties of the procedures are analyzed. Illustrations are given in the study of the emphasis placed on historical data used in the forecasting process, for each of the models \(M_ T\), \(S_ T\) and \(EM_ T\). Finally, results of fitting these models to two real sets of data are given. As part of this analysis, forecasts are made over the last 5 time periods of each of the data sets for each of the three models.