東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

A Study of Log-concave Mixture Models.

Hu, Hao.

Linked to FindBook

Google Book

Amazon

博客來

A Study of Log-concave Mixture Models.

Record Type:	Electronic resources : Monograph/item
Title/Author:	A Study of Log-concave Mixture Models./
Author:	Hu, Hao.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2016,
Description:	122 p.
Notes:	Source: Dissertation Abstracts International, Volume: 78-08(E), Section: B.
Contained By:	Dissertation Abstracts International78-08B(E).
Subject:	Statistics. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10583428
ISBN:	9781369621075

A Study of Log-concave Mixture Models.
Hu, Hao.

A Study of Log-concave Mixture Models. - Ann Arbor : ProQuest Dissertations & Theses, 2016 - 122 p.

Source: Dissertation Abstracts International, Volume: 78-08(E), Section: B.

Thesis (Ph.D.)--North Carolina State University, 2016.

Mixture models are widely used when data are from a number of different components. Traditional parametric mixture models can be estimated via the expectationmaximization algorithm (known as the EM-algorithm) based on their parametric assumptions. However, these assumptions are sometimes too restrictive and the estimation results are biased if the models are misspecified. To relax the parametric assumption, we apply a log-concave shape constraint. This dissertation analyzes the log-concave mixture models, which are more exible and general than the traditional parametric mixture models. We developed a nonparametric log-concave maximum likelihood estimator (LCMLE) for the log-concave mixture model. In particular, we investigate the theoretical properties, computational algorithms and applications in clustering. We also develop the computational algorithms for the logconcave mixtures of regression model and its extension.

ISBN: 9781369621075Subjects--Topical Terms:

517247
Statistics.

A Study of Log-concave Mixture Models.
LDR:01787nmm a2200277 4500 001 2128507
005 20180104132948.5
008 180830s2016 ||||||||||||||||| ||eng d
020 $a 9781369621075
035 $a (MiAaPQ)AAI10583428
035 $a AAI10583428
040 $a MiAaPQ $c MiAaPQ
100 1 $a Hu, Hao. $3 1620319
245 1 2 $a A Study of Log-concave Mixture Models.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2016
300 $a 122 p.
500 $a Source: Dissertation Abstracts International, Volume: 78-08(E), Section: B.
500 $a Adviser: Yichao Wu.
502 $a Thesis (Ph.D.)--North Carolina State University, 2016.
520 $a Mixture models are widely used when data are from a number of different components. Traditional parametric mixture models can be estimated via the expectationmaximization algorithm (known as the EM-algorithm) based on their parametric assumptions. However, these assumptions are sometimes too restrictive and the estimation results are biased if the models are misspecified. To relax the parametric assumption, we apply a log-concave shape constraint. This dissertation analyzes the log-concave mixture models, which are more exible and general than the traditional parametric mixture models. We developed a nonparametric log-concave maximum likelihood estimator (LCMLE) for the log-concave mixture model. In particular, we investigate the theoretical properties, computational algorithms and applications in clustering. We also develop the computational algorithms for the logconcave mixtures of regression model and its extension.
590 $a School code: 0155.
650 4 $a Statistics. $3 517247
690 $a 0463
710 2 $a North Carolina State University. $3 1018772
773 0 $t Dissertation Abstracts International $g 78-08B(E).
790 $a 0155
791 $a Ph.D.
792 $a 2016
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10583428