Modelling autoregressive processes with a shifting mean

This paper contains a nonlinear, nonstationary autoregressive model whose intercept changes deterministically over time. The intercept is a flexible function of time, and its construction bears some resemblance to neural network models. A modeling techniq

Detalles Bibliográficos
Autores Principales: Terasvirta, Timo, González-Gómez, Andrés
Formato: Documento de trabajo (Working Paper)
Lenguaje:Español (Spanish)
Publicado: Banco de la República 2006
Materias:
Acceso en línea:http://repositorio.banrep.gov.co/handle/20.500.12134/5438
Descripción
Sumario:This paper contains a nonlinear, nonstationary autoregressive model whose intercept changes deterministically over time. The intercept is a flexible function of time, and its construction bears some resemblance to neural network models. A modeling techniq