Maximum likelihood estimation of the multivariate normal mixture model

O. Boldea, J.R. Magnus

Research output: Contribution to journalArticleScientificpeer-review

203 Downloads (Pure)

Abstract

The Hessian of the multivariate normal mixture model is derived, and estimators of the information matrix are obtained, thus enabling consistent estimation of all parameters and their precisions. The usefulness of the new theory is illustrated with two examples and some simulation experiments. The newly proposed estimators appear to be superior to the existing ones.
Original languageEnglish
Pages (from-to)1539-1549
JournalJournal of the American Statistical Association
Volume104
Issue number488
Publication statusPublished - 2009

Fingerprint

Normal Mixture
Multivariate Normal
Mixture Model
Maximum Likelihood Estimation
Estimator
Consistent Estimation
Information Matrix
Simulation Experiment
Maximum likelihood estimation
Mixture model
Usefulness
Simulation experiment

Cite this

@article{c5d9a58c6bc24098bfedd9382b099a18,
title = "Maximum likelihood estimation of the multivariate normal mixture model",
abstract = "The Hessian of the multivariate normal mixture model is derived, and estimators of the information matrix are obtained, thus enabling consistent estimation of all parameters and their precisions. The usefulness of the new theory is illustrated with two examples and some simulation experiments. The newly proposed estimators appear to be superior to the existing ones.",
author = "O. Boldea and J.R. Magnus",
year = "2009",
language = "English",
volume = "104",
pages = "1539--1549",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "488",

}

Maximum likelihood estimation of the multivariate normal mixture model. / Boldea, O.; Magnus, J.R.

In: Journal of the American Statistical Association, Vol. 104, No. 488, 2009, p. 1539-1549.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Maximum likelihood estimation of the multivariate normal mixture model

AU - Boldea, O.

AU - Magnus, J.R.

PY - 2009

Y1 - 2009

N2 - The Hessian of the multivariate normal mixture model is derived, and estimators of the information matrix are obtained, thus enabling consistent estimation of all parameters and their precisions. The usefulness of the new theory is illustrated with two examples and some simulation experiments. The newly proposed estimators appear to be superior to the existing ones.

AB - The Hessian of the multivariate normal mixture model is derived, and estimators of the information matrix are obtained, thus enabling consistent estimation of all parameters and their precisions. The usefulness of the new theory is illustrated with two examples and some simulation experiments. The newly proposed estimators appear to be superior to the existing ones.

M3 - Article

VL - 104

SP - 1539

EP - 1549

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 488

ER -