Blackwell Reference Online

Extract

There are numerous statistical approaches to forecasting, from simple, regression-based methods to optimal statistical procedures formulated in stochastic statespace terms. Since it would be impossible to review all these methods here, the present chapter tries to distill, from this large mixture of models and methods, those that the authors feel have most significance in theoretical and practical terms within the specific context of economic forecasting. Most of the statistical forecasting methods referred to in the chapter are model-based, in the sense that the forecasting operation is carried out subsequent to the statistical identification and estimation of a suitable (usually stochastic) mathematical model based on the available time-series data. Consequently, in the subsequent subsections, the differentiation between forecasting methods is based largely on the type of model used to characterize the data. The forecasting procedures themselves are simply devices for utilizing the model to project its output forward into the future in stochastic terms, normally through the evolution of the mean and standard error bands associated with forecast distributions. One of the best known statistical approaches to forecasting derives from classical regression analysis. This approach has been used by most scientific disciplines and has served as the basis for a wide range of subsequent ... log in or subscribe to read full text