Determine the number of components in a mixture model by the extended KS test

Authors:
Ming-Heng Zhang;Qian-Sheng Cheng
Affiliations:
Research School of Economics, Shanghai University of Finance and Economics, Shanghai 200433, China;School of Mathematical Sciences, Peking University, Beijing 100871, China
Venue:
Pattern Recognition Letters
Year:
2004

Citing 5
Cited 1

The nature of statistical learning theory

The nature of statistical learning theory
On cross validation for model selection

Neural Computation
Computing the cumulative distribution function of the Kolmogorov-Smirnov statistic

Computational Statistics & Data Analysis
Assessing a Mixture Model for Clustering with the Integrated Completed Likelihood

IEEE Transactions on Pattern Analysis and Machine Intelligence
On convergence properties of the em algorithm for gaussian mixtures

Neural Computation

Mutual Information Analysis: How, When and Why?

CHES '09 Proceedings of the 11th International Workshop on Cryptographic Hardware and Embedded Systems

Quantified Score

Hi-index	0.10

Visualization

Abstract

In this paper, an extended KS test is proposed to determine the number of components in a mixture model. An approach based on the EM procedure is constructed by the extended KS test in parallel to do the computation. The simulation studies are given to illustrate the method proposed.