A problem frequently encountered in the analysis of time series data is that it has interventions. Intervention, a kind of impact from exterior, has a various forms of effects. Box and Tiao developed intervention analysis to reflect more practical eff...
A problem frequently encountered in the analysis of time series data is that it has interventions. Intervention, a kind of impact from exterior, has a various forms of effects. Box and Tiao developed intervention analysis to reflect more practical effects of interventions. When interventions occur, it calls for an urgent need of finding out the intervention points.
In the case of analysing the overall data with ignoring the intervention points, the accurate characteristic of the data can not be drawn by a finite number of parameters. The aim of this paper is to estimate unknown intervention points. This paper proposes two estimation procedures and presents the result of pragmatic examples which illustrate the usefulness of these two procedures in estimating unknown intervention points.