Forecasting Short Seasonal Time Series Using Aggregate and Analogous Series

In forecasting short seasonal time series, Michael Leonard recommends seeking additional sources of information outside the (sparse) historical data on the items themselves. Two potentially promising sources of external information are the histories of the analogous products and of product aggregate in which the item is a component. The article describes a selection tree that can assist the forecaster in choosing an appropriate modeling approach. Copyright International Institute of Forecasters, 2007