Bayesian nonparametric spectral analysis of locally stationary processes
From MaRDI portal
Publication:5984576
DOI10.48550/ARXIV.2303.11561arXiv2303.11561MaRDI QIDQ5984576FDOQ5984576
Authors: Yifu Tang, Claudia Kirch, Jeong Eun Lee, Renate Meyer 
Publication date: 21 March 2023
Abstract: Based on a novel dynamic Whittle likelihood approximation for locally stationary processes, a Bayesian nonparametric approach to estimating the time-varying spectral density is proposed. This dynamic frequency-domain based likelihood approximation is able to depict the time-frequency evolution of the process by utilizing the moving periodogram previously introduced in the bootstrap literature. The posterior distribution is obtained by updating a bivariate extension of the Bernstein-Dirichlet process prior with the dynamic Whittle likelihood. Asymptotic properties such as sup-norm posterior consistency and L2-norm posterior contraction rates are presented. Additionally, this methodology enables model selection between stationarity and non-stationarity based on the Bayes factor. The finite-sample performance of the method is investigated in simulation studies and applications to real-life data-sets are presented.
Cited In (1)
This page was built for publication: Bayesian nonparametric spectral analysis of locally stationary processes
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5984576)
