Some Aspects of Applied Forecasting Methods. A Modelling Approach

      Аннотация: This book has brought out inferential methods to forecasting with linear statistical and time series models, the various forecasting methods existing in the literature have been briefly reviewed with inferential problems on them. In view of the importance of forecasting is empirical research, some new procedures for applied forecasting have been developed.Here, these techniques are developed by using Internally Studentized Residuals. Further, a modified Box-Jenkins methodology has been presented for auto Integrated Moving average model ARIMA(p,d,q) based on Internally Studentized Residuals. Under Diagnostic checking, a modified L Jung and Box statistic for testing the residuals has been proposed. The forecasts to be obtained from this methodology may be used as benchmark to compare with forecasts to be yielded by other forecasting techniques.

Ключевые слова: Statistics