EM Algorithm for Normal Mixture Likelihoods

Chukiat Viwatwongkasem

doi:10.1109/IEECON.2018.8712275

EM Algorithm for Normal Mixture Likelihoods

Chukiat Viwatwongkasem

Department of Biostatistics

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Citations (Scopus)

Abstract

The purpose of the study is to use the expectation-maximization (EM) algorithm for finding the maximum likelihood estimates (MLEs) under the normal mixture models in which allow heterogeneity in forms of the multi-nodes, skewed, long-tailed, and/or contaminated distributions. The motivational application of the standardized morbidity ratio (SMR) of geographical HIV/AIDS data displaying on a map among all study provinces in Thailand 2013 is illustrated. The results showed that the normal mixture model fitted data well with the nice MLEs corresponding to the EM algorithm coping with good yielding both numerically stable convergence and the fine estimates of local maximum points. Another advantage of EM algorithm was in adding up the latent unobserved probabilities of each study province belonging to the component of normal mixture in solving the problem of the incomplete data while other algorithms, such as Newton-Raphson and Fisher Scoring, couldn't be able to augment those unobserved missing data. However, EM algorithm seemed to have slow convergence.

Original language	English
Title of host publication	iEECON 2018 - 6th International Electrical Engineering Congress
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781538623176
DOIs	https://doi.org/10.1109/IEECON.2018.8712275
Publication status	Published - 2 Jul 2018
Event	6th International Electrical Engineering Congress, iEECON 2018 - Krabi, Thailand Duration: 7 Mar 2018 → 9 Mar 2018

Publication series

Name	iEECON 2018 - 6th International Electrical Engineering Congress

Conference

Conference	6th International Electrical Engineering Congress, iEECON 2018
Country/Territory	Thailand
City	Krabi
Period	7/03/18 → 9/03/18

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

EM algorithm
HIV/AIDS
geographical mapping
normal mixture model
standardized morbidity/mortality ratio

Access to Document

10.1109/IEECON.2018.8712275

Cite this

@inproceedings{d96aab45f35744dea7acb3cb4bbba856,

title = "EM Algorithm for Normal Mixture Likelihoods",

abstract = "The purpose of the study is to use the expectation-maximization (EM) algorithm for finding the maximum likelihood estimates (MLEs) under the normal mixture models in which allow heterogeneity in forms of the multi-nodes, skewed, long-tailed, and/or contaminated distributions. The motivational application of the standardized morbidity ratio (SMR) of geographical HIV/AIDS data displaying on a map among all study provinces in Thailand 2013 is illustrated. The results showed that the normal mixture model fitted data well with the nice MLEs corresponding to the EM algorithm coping with good yielding both numerically stable convergence and the fine estimates of local maximum points. Another advantage of EM algorithm was in adding up the latent unobserved probabilities of each study province belonging to the component of normal mixture in solving the problem of the incomplete data while other algorithms, such as Newton-Raphson and Fisher Scoring, couldn't be able to augment those unobserved missing data. However, EM algorithm seemed to have slow convergence.",

keywords = "EM algorithm, HIV/AIDS, geographical mapping, normal mixture model, standardized morbidity/mortality ratio",

author = "Chukiat Viwatwongkasem",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 6th International Electrical Engineering Congress, iEECON 2018 ; Conference date: 07-03-2018 Through 09-03-2018",

year = "2018",

month = jul,

day = "2",

doi = "10.1109/IEECON.2018.8712275",

language = "English",

series = "iEECON 2018 - 6th International Electrical Engineering Congress",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "iEECON 2018 - 6th International Electrical Engineering Congress",

}

Viwatwongkasem, C 2018, EM Algorithm for Normal Mixture Likelihoods. in iEECON 2018 - 6th International Electrical Engineering Congress., 8712275, iEECON 2018 - 6th International Electrical Engineering Congress, Institute of Electrical and Electronics Engineers Inc., 6th International Electrical Engineering Congress, iEECON 2018, Krabi, Thailand, 7/03/18. https://doi.org/10.1109/IEECON.2018.8712275

EM Algorithm for Normal Mixture Likelihoods. / Viwatwongkasem, Chukiat.
iEECON 2018 - 6th International Electrical Engineering Congress. Institute of Electrical and Electronics Engineers Inc., 2018. 8712275 (iEECON 2018 - 6th International Electrical Engineering Congress).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - EM Algorithm for Normal Mixture Likelihoods

AU - Viwatwongkasem, Chukiat

PY - 2018/7/2

Y1 - 2018/7/2

N2 - The purpose of the study is to use the expectation-maximization (EM) algorithm for finding the maximum likelihood estimates (MLEs) under the normal mixture models in which allow heterogeneity in forms of the multi-nodes, skewed, long-tailed, and/or contaminated distributions. The motivational application of the standardized morbidity ratio (SMR) of geographical HIV/AIDS data displaying on a map among all study provinces in Thailand 2013 is illustrated. The results showed that the normal mixture model fitted data well with the nice MLEs corresponding to the EM algorithm coping with good yielding both numerically stable convergence and the fine estimates of local maximum points. Another advantage of EM algorithm was in adding up the latent unobserved probabilities of each study province belonging to the component of normal mixture in solving the problem of the incomplete data while other algorithms, such as Newton-Raphson and Fisher Scoring, couldn't be able to augment those unobserved missing data. However, EM algorithm seemed to have slow convergence.

AB - The purpose of the study is to use the expectation-maximization (EM) algorithm for finding the maximum likelihood estimates (MLEs) under the normal mixture models in which allow heterogeneity in forms of the multi-nodes, skewed, long-tailed, and/or contaminated distributions. The motivational application of the standardized morbidity ratio (SMR) of geographical HIV/AIDS data displaying on a map among all study provinces in Thailand 2013 is illustrated. The results showed that the normal mixture model fitted data well with the nice MLEs corresponding to the EM algorithm coping with good yielding both numerically stable convergence and the fine estimates of local maximum points. Another advantage of EM algorithm was in adding up the latent unobserved probabilities of each study province belonging to the component of normal mixture in solving the problem of the incomplete data while other algorithms, such as Newton-Raphson and Fisher Scoring, couldn't be able to augment those unobserved missing data. However, EM algorithm seemed to have slow convergence.

KW - EM algorithm

KW - HIV/AIDS

KW - geographical mapping

KW - normal mixture model

KW - standardized morbidity/mortality ratio

UR - https://www.scopus.com/pages/publications/85066636908

U2 - 10.1109/IEECON.2018.8712275

DO - 10.1109/IEECON.2018.8712275

M3 - Conference contribution

AN - SCOPUS:85066636908

T3 - iEECON 2018 - 6th International Electrical Engineering Congress

BT - iEECON 2018 - 6th International Electrical Engineering Congress

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 6th International Electrical Engineering Congress, iEECON 2018

Y2 - 7 March 2018 through 9 March 2018

ER -

EM Algorithm for Normal Mixture Likelihoods

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this