Home > Research > Publications & Outputs > Filtering Methods for Mixture Models .

Research

Electronic data

Links

Keywords

View graph of relations

Filtering Methods for Mixture Models .

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Filtering Methods for Mixture Models . / Fearnhead, P; Meligkotsidou, L.
In: Journal of Computational and Graphical Statistics, Vol. 16, No. 3, 2007, p. 586-607.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Fearnhead, P & Meligkotsidou, L 2007, 'Filtering Methods for Mixture Models .', Journal of Computational and Graphical Statistics, vol. 16, no. 3, pp. 586-607. <http://lysander.asa.catchword.org/vl=1418380/cl=27/nw=1/rpsv/cw/asa/10618600/v16n3/s4/p586>

APA

Fearnhead, P., & Meligkotsidou, L. (2007). Filtering Methods for Mixture Models . Journal of Computational and Graphical Statistics, 16(3), 586-607. http://lysander.asa.catchword.org/vl=1418380/cl=27/nw=1/rpsv/cw/asa/10618600/v16n3/s4/p586

Vancouver

Fearnhead P, Meligkotsidou L. Filtering Methods for Mixture Models . Journal of Computational and Graphical Statistics. 2007;16(3):586-607.

Author

Fearnhead, P ; Meligkotsidou, L. / Filtering Methods for Mixture Models . In: Journal of Computational and Graphical Statistics. 2007 ; Vol. 16, No. 3. pp. 586-607.

Bibtex

@article{aa95431ccc3e47258b97fd79d70abbc0,
title = "Filtering Methods for Mixture Models .",
abstract = "We consider Bayesian inference for mixture distributions of known number of components via a set of filtering recursions. We extend a method - proposed in an earlier article - of direct simulation for discrete mixture distributions in order to analyze continuous mixture models. Furthermore, we introduce resampling steps similar to those in particle filters within the steps of the filtering recursions, which make calculations efficient and enable us to analyze larger datasets. The proposed algorithm for {"}resampled direct simulation{"} is a generalization of the particle filter which allows for merging identical/similar particles prior to resampling. We compare the proposed algorithm with this particle filter and with the Gibbs sampler using simulated data and real datasets.",
keywords = "Direct Simulation, Gibbs Sampling, Importance Sampling, Particle Filters, Perfect Simulation, Rejection Sampling",
author = "P Fearnhead and L Meligkotsidou",
year = "2007",
language = "English",
volume = "16",
pages = "586--607",
journal = "Journal of Computational and Graphical Statistics",
issn = "1537-2715",
publisher = "American Statistical Association",
number = "3",

}

RIS

TY - JOUR

T1 - Filtering Methods for Mixture Models .

AU - Fearnhead, P

AU - Meligkotsidou, L

PY - 2007

Y1 - 2007

N2 - We consider Bayesian inference for mixture distributions of known number of components via a set of filtering recursions. We extend a method - proposed in an earlier article - of direct simulation for discrete mixture distributions in order to analyze continuous mixture models. Furthermore, we introduce resampling steps similar to those in particle filters within the steps of the filtering recursions, which make calculations efficient and enable us to analyze larger datasets. The proposed algorithm for "resampled direct simulation" is a generalization of the particle filter which allows for merging identical/similar particles prior to resampling. We compare the proposed algorithm with this particle filter and with the Gibbs sampler using simulated data and real datasets.

AB - We consider Bayesian inference for mixture distributions of known number of components via a set of filtering recursions. We extend a method - proposed in an earlier article - of direct simulation for discrete mixture distributions in order to analyze continuous mixture models. Furthermore, we introduce resampling steps similar to those in particle filters within the steps of the filtering recursions, which make calculations efficient and enable us to analyze larger datasets. The proposed algorithm for "resampled direct simulation" is a generalization of the particle filter which allows for merging identical/similar particles prior to resampling. We compare the proposed algorithm with this particle filter and with the Gibbs sampler using simulated data and real datasets.

KW - Direct Simulation

KW - Gibbs Sampling

KW - Importance Sampling

KW - Particle Filters

KW - Perfect Simulation

KW - Rejection Sampling

M3 - Journal article

VL - 16

SP - 586

EP - 607

JO - Journal of Computational and Graphical Statistics

JF - Journal of Computational and Graphical Statistics

SN - 1537-2715

IS - 3

ER -