haku: @indexterm STRUCTURAL CHANGE / yhteensä: 382
viite: 81 / 382
Tekijä: | Koopman, S. J. Shephard, N. |
Otsikko: | Detecting shocks: Ouliers and breaks in time series. |
Lehti: | Journal of Econometrics
1997 : OCT, VOL. 80:2, p. 387-422 |
Asiasana: | MONTE CARLO TECHNIQUE TIME SERIES SMALL BUSINESS REGRESSION ANALYSIS STRUCTURAL CHANGE |
Kieli: | eng |
Tiivistelmä: | The purpose of this paper is to develop powerful methods detecting specific inadequacies in time-series models. The authors use the unoubserved component or structural approach to time-series modelling, that is dynamic linear methods, to provide procedures which have much in common with the well- known diagnostics for multiple-regression models. The procedures, illustrated with four examples, permit keen insights into the fragidity of inferences to specific shocks, such as outliers and level breaks. Although the emphasis is mostly on parameter estimation, forecasts are also considered. |
SCIMA