Publications
Detailed Information
Bayesian analysis of multivariate mixture models via factor analyzer : 인자분해를 통한 다변량 혼합 모형의 베이지안 분석
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 김용대 | - |
dc.contributor.author | 김재석 | - |
dc.date.accessioned | 2017-07-14T00:30:33Z | - |
dc.date.available | 2017-07-14T00:30:33Z | - |
dc.date.issued | 2012-08 | - |
dc.identifier.other | 000000004821 | - |
dc.identifier.uri | https://hdl.handle.net/10371/121137 | - |
dc.description | 학위논문 (박사)-- 서울대학교 대학원 : 통계학과, 2012. 8. 김용대. | - |
dc.description.abstract | We consider a new Bayesian finite mixture model for multivariate data.
A problem is to estimate the covariance matrix since the number of parameters for the covariance matrix is squarely proportional to the dimension of data. Also, the inverting large dimensional covariate, which is necessary for MCMC algorithms, is very time consuming and practically almost prohibited. In this thesis, we propose a way of reducing the parameters in the covariance matrices by use of the factor model. That is, the dependence structure of each component is assumed to be represented by linear combinations of factors. To simply the model and improve interpretability further, we allow some factors can be shared across the components. From numerical studies, we confirmed that our method well perform with different component covariance structure. | - |
dc.description.tableofcontents | 1 Introduction
1.1 Overview 1.2 Outline of the thesis 2 Reviews 2.1 Univariate case 2.1.1 EM algorithm 2.1.2 Bayesian method 2.2 Multivariate case 2.3 Mixture of factor analyzers 3 Bayesian factor clustering 3.1 Model 3.2 Priors 3.3 RJMCMC 4 Numerical studies 4.1 Simulation studies 4.1.1 Simulation 1(Covariance matrix structures) 4.1.2 Simulation 2(The number of cluster) 4.1.3 Simulation 3(The dimension) 4.1.4 Simulation 4(The number of observations) 4.1.5 Simulation 5(The unbalanced data) 4.1.6 Simulation 6(The hyper parameter of MCRP) 4.2 Real data analysis 5 Concluding remarks | - |
dc.format | application/pdf | - |
dc.format.extent | 999698 bytes | - |
dc.format.medium | application/pdf | - |
dc.language.iso | en | - |
dc.publisher | 서울대학교 대학원 | - |
dc.subject | Bayesian | - |
dc.subject | Mixture model | - |
dc.subject | Clustering | - |
dc.subject | Factor analysis | - |
dc.subject | RJMCMC | - |
dc.title | Bayesian analysis of multivariate mixture models via factor analyzer | - |
dc.title.alternative | 인자분해를 통한 다변량 혼합 모형의 베이지안 분석 | - |
dc.type | Thesis | - |
dc.contributor.AlternativeAuthor | Jaeseok Kim | - |
dc.description.degree | Doctor | - |
dc.citation.pages | LIX, 59 | - |
dc.contributor.affiliation | 자연과학대학 통계학과 | - |
dc.date.awarded | 2012-08 | - |
- Appears in Collections:
- Files in This Item:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.