Identification

Abstract

A fundamental difficulty in statistical analysis is the choice of an appropriate model. This is particularly pronounced in time series analysis. The book by Box and Jenkins (1970) has been successful in popularizing linear time series models through the formulation of an iterative process of model building, consisting of the stages of identification, estimation and diagnostic checking. This approach is commonly referred to as the Box-Jenkins approach. It seems that the general philosophy of this approach is to allow the modeller a fair degree of flexibility in exercising his subjective judgement as to which one of several candidate models he may adopt. The approach has been popular and, in the hands of experienced time series analysts, many successful results have been reported. Their emphasis on diagnostic checking is particularly relevant in the present context.