Bayesian estimation of parameters of a damped sinusoidal model by a Markov chain Monte Carlo method

Author

Barone, Piero ; Ragona, Raffaello

Author_Institution

Ist. per le Applicazioni del Calcolo, CNR, Rome, Italy

Volume

45

Issue

7

fYear

1997

fDate

7/1/1997 12:00:00 AM

Firstpage

1806

Lastpage

1814

Abstract

A dynamic Monte Carlo method is proposed to compute the posterior means and covariances of the parameters of a damped sinusoidal model when an informative prior distribution is known. The Bayesian framework provides a sound mathematical ground, which possibly allows one to overcome the approximations commonly used to cope with this difficult problem. Some simulations results are provided, which support the conclusion that the prior information can also be significantly improved when the data have a low signal-to-noise ratio

Keywords

Bayes methods; Markov processes; Monte Carlo methods; covariance analysis; parameter estimation; spectral analysis; Bayesian estimation; Markov chain Monte Carlo method; covariance; damped sinusoidal model; dynamic Monte Carlo metho; informative prior distribution; parameters; posterior mean; signal-to-noise ratio; Acoustic noise; Bayesian methods; Computational modeling; Data models; Differential equations; Distributed computing; Helium; Parameter estimation; Signal to noise ratio; Statistics;

fLanguage

English

Journal_Title

Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

1053-587X

Type

jour

DOI

10.1109/78.599950

Filename

599950