Autori
Liseo, Brunero
Marinucci, Domenico
Petrella, Lea

Titolo
Bayesian semiparametric inference on long range dependence
Periodico
Università degli Studi di Roma "La Sapienza" - Dipartimento di Studi Geoeconomici, Linguistici, Statistici e Storici per l'Analisi regionale. Working papers
Anno: 1999 - Fascicolo: 8 - Pagina iniziale: 1 - Pagina finale: 24

We develop a Bayesian semiparametric procedure for the anlysis of stationary long range dependent time series. We use frequency domain methods to partition the infinite-dimensional parameter space into regions where genuine prior information on the form of the spectral density is available, and others where vague prior beliefs are adopted; the solution of the partition problem, which is equivalent to bandwidth choise from a frequentist point of view, is obtained via Bayesian model selection techniques. We then derive a tight characterization of the class of admissible noninformative priors for nonparametric inference on the spectral density of a stationary process. Some asymptotic properties of our technique and comparisons with frequentist approaches are also considered; the suggested procedure is finally implemented via MCMC methods in a Monte Carlo experiment and on real data




Esportazione dati in Refworks (solo per utenti abilitati)