Bayesian nonparametric estimation of the spectral density of a long or intermediate memory Gaussian process

Rousseau, Judith; Chopin, Nicolas; Liseo, Brunero

Rousseau, Judith; Chopin, Nicolas; Liseo, Brunero (2012), Bayesian nonparametric estimation of the spectral density of a long or intermediate memory Gaussian process, Annals of Statistics, 40, 2, p. 964-995. http://dx.doi.org/10.1214/11-AOS955SUPP

Type

Article accepté pour publication ou publié

External document link

http://hal.archives-ouvertes.fr/hal-00504969/fr/

Date

2012

Journal name

Annals of Statistics

Volume

Number

Publisher

Institute of Mathematical Statistics

Pages

964-995

Abstract (EN)

A stationary Gaussian process is said to be long-range dependent (resp., anti-persistent) if its spectral density f(λ) can be written as f(λ)=|λ|−2dg(|λ|), where 0

Subjects / Keywords

rates of convergence; Gaussian long memory processes; FEXP priors; consistency; Bayesian nonparametric

JEL

C14 - Semiparametric and Nonparametric Methods: General
C11 - Bayesian Analysis: General