Prediction intervals for nearly nonstationary AR(1)-processes

Ferretti, NélidaRomo, JuanUniversidad Carlos III de Madrid. Departamento de Estadística2009-02-202009-02-201993-10https://hdl.handle.net/10016/3735We construct prediction intervals for the observations of first-order autoregressive processes when the model approaches a nonstationary situation with a unit root. The intervals that we propose contain an s-step future value with a given asymptotic probability conditional on the observation. A simulation study has been also carried out to illustrate our results.application/pdfengAtribución-NoComercial-SinDerivadas 3.0 EspañaPrediction intervalsNearly nonstationary time seriesCoverage probabilityPrediction intervals for nearly nonstationary AR(1)-processesworking paperEstadísticaopen access