Home > Research > Publications & Outputs > An EM-type algorithm for multivariate mixture m...

Research

Links

Text available via DOI:

Keywords

View graph of relations

An EM-type algorithm for multivariate mixture models.

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

Published

Standard

An EM-type algorithm for multivariate mixture models. / Oskrochi, Gholam; Davies, R. B.
In: Statistics and Computing, Vol. 7, No. 2, 06.1997, p. 145-151.

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

Harvard

Oskrochi, G & Davies, RB 1997, 'An EM-type algorithm for multivariate mixture models.', Statistics and Computing, vol. 7, no. 2, pp. 145-151. https://doi.org/10.1023/A:1018525800226

APA

Oskrochi, G., & Davies, R. B. (1997). An EM-type algorithm for multivariate mixture models. Statistics and Computing, 7(2), 145-151. https://doi.org/10.1023/A:1018525800226

Vancouver

Oskrochi G, Davies RB. An EM-type algorithm for multivariate mixture models. Statistics and Computing. 1997 Jun;7(2):145-151. doi: 10.1023/A:1018525800226

Author

Oskrochi, Gholam ; Davies, R. B. / An EM-type algorithm for multivariate mixture models. In: Statistics and Computing. 1997 ; Vol. 7, No. 2. pp. 145-151.

Bibtex

@article{9b45708ac5be463d8ce068b9aff339fa,
title = "An EM-type algorithm for multivariate mixture models.",
abstract = "This paper introduces a new approach, based on dependent univariate GLMs, for fitting multivariate mixture models. This approach is a multivariate generalization of the method for univariate mixtures presented by Hinde (1982). Its accuracy and efficiency are compared with direct maximization of the log-likelihood. Using a simulation study, we also compare the efficiency of Monte Carlo and Gaussian quadrature methods for approximating the mixture distribution. The new approach with Gaussian quadrature outperforms the alternative methods considered. The work is motivated by the multivariate mixture models which have been proposed for modelling changes of employment states at an individual level. Similar formulations are of interest for modelling movement between other social and economic states and multivariate mixture models also occur in biostatistics and epidemiology.",
keywords = "Multivariate generalized linear models - Markov model - EM algorithm - random effect models - Monte Carlo simulation - Cholesky decomposition",
author = "Gholam Oskrochi and Davies, {R. B.}",
year = "1997",
month = jun,
doi = "10.1023/A:1018525800226",
language = "English",
volume = "7",
pages = "145--151",
journal = "Statistics and Computing",
issn = "0960-3174",
publisher = "Springer Netherlands",
number = "2",

}

RIS

TY - JOUR

T1 - An EM-type algorithm for multivariate mixture models.

AU - Oskrochi, Gholam

AU - Davies, R. B.

PY - 1997/6

Y1 - 1997/6

N2 - This paper introduces a new approach, based on dependent univariate GLMs, for fitting multivariate mixture models. This approach is a multivariate generalization of the method for univariate mixtures presented by Hinde (1982). Its accuracy and efficiency are compared with direct maximization of the log-likelihood. Using a simulation study, we also compare the efficiency of Monte Carlo and Gaussian quadrature methods for approximating the mixture distribution. The new approach with Gaussian quadrature outperforms the alternative methods considered. The work is motivated by the multivariate mixture models which have been proposed for modelling changes of employment states at an individual level. Similar formulations are of interest for modelling movement between other social and economic states and multivariate mixture models also occur in biostatistics and epidemiology.

AB - This paper introduces a new approach, based on dependent univariate GLMs, for fitting multivariate mixture models. This approach is a multivariate generalization of the method for univariate mixtures presented by Hinde (1982). Its accuracy and efficiency are compared with direct maximization of the log-likelihood. Using a simulation study, we also compare the efficiency of Monte Carlo and Gaussian quadrature methods for approximating the mixture distribution. The new approach with Gaussian quadrature outperforms the alternative methods considered. The work is motivated by the multivariate mixture models which have been proposed for modelling changes of employment states at an individual level. Similar formulations are of interest for modelling movement between other social and economic states and multivariate mixture models also occur in biostatistics and epidemiology.

KW - Multivariate generalized linear models - Markov model - EM algorithm - random effect models - Monte Carlo simulation - Cholesky decomposition

U2 - 10.1023/A:1018525800226

DO - 10.1023/A:1018525800226

M3 - Journal article

VL - 7

SP - 145

EP - 151

JO - Statistics and Computing

JF - Statistics and Computing

SN - 0960-3174

IS - 2

ER -