An effective method for selecting the number of components in density mixtures

Abstract

This paper proposes a simple and effective method for selecting the number of components in density mixtures. The proposed method is based on a combination of Stein's unbiased risk estimation and the plug-in principle. Also provided by this paper are numerical results comparing several popular methods for Gaussian mixture estimation. These results demonstrate the superior performance of the proposed method.

Keywords:

• Bootstrap
• Density mixtures
• Gaussian mixtures
• Mixture distributions
• Plug-in principle
• Unbiased risk estimation

Acknowledgements

The authors are most grateful to the referee for his/her constructive comments and suggestions, most of which were incorporated into the final version of the paper. The authors would also like to thank Professor Peter Green for making the N mix code of Richardson and Green (1997) available.