We study the problem of intervention e ects generating various types of outliers in a rst order integer valued autoregressive model with Poisson innovations.
We concentrate on outliers which enter the dynamics and can be seen as e ects of extraordinary events. We consider three di erent scenarios, namely the detection of an intervention e ect of a known type at a known time, the detection of an intervention e ect of unknown type at a known time and the detection of an intervention e ect when the type and the time are both unknown. We develop F-tests and score tests for the rst scenario. For the second and third scenarios we rely on the maximum of the di erent F-type or score statistics. The usefulness of the proposed approach is illustrated using simulated examples.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados