[Home ] [Archive]   [ فارسی ]  
:: Main :: About :: Current Issue :: Archive :: Search :: Submit :: Contact ::
:: Volume 24, Issue 2 (3-2014) ::
مجله‌ی بررسی‌ها 2014, 24(2): 145-165 Back to browse issues page
Four Methods of Estimating Parameters of a Gaussian Mixture Model
Donya Rahmani, Adel Mohammadpour *
Abstract:   (3596 Views)

Gaussian mixture model is the most widely used finite mixture model. The important feature of this model is its flexibility with respect to various forms of continuous distributions. Because the most important part of model fitting is the estimation of its parameters, we want to estimate two-component Gaussian mixture model parameters through the four estimation methods. We introduce a two-component Gaussian mixture model and then estimate model parameters using the method of moment and maximum likelihood by the name of analytical and numerical solutions, respectively. In continuation, the parameters are estimated using the EM algorithm and the Gibbs sampler algorithm. In conclusion, the results of these methods are compared with each other. We try to solve an estimation problem with the four common methods and introduce advantage and disadvantage of them for users.

Keywords: Analytical and numerical solution, EM algorithm, Gibbs sampler, Gaussian mixture model.
Full-Text [PDF 246 kb]   (5181 Downloads)    
Type of Study: Research | Subject: General
Received: 2011/10/31 | Accepted: 2014/11/24 | Published: 2015/12/6
Send email to the article author

Add your comments about this article
Your username or Email:

CAPTCHA


XML   Persian Abstract   Print


Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

Rahmani D, Mohammadpour A. Four Methods of Estimating Parameters of a Gaussian Mixture Model . مجله‌ی بررسی‌ها. 2014; 24 (2) :145-165
URL: http://ijoss.srtc.ac.ir/article-1-30-en.html


Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Volume 24, Issue 2 (3-2014) Back to browse issues page
مجله‌ی بررسی‌های آمار رسمی ایران (علمی - ترویجی) Ijoss Iranian Journal of Official Statistics Studies
Persian site map - English site map - Created in 0.04 seconds with 30 queries by YEKTAWEB 4341