Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
10526246 | Statistics & Probability Letters | 2005 | 10 Pages |
Abstract
We present two results that pertain to the choosing of the centering distribution in a Bayesian setup with a Dirichlet process mixture prior based on Gaussian kernels. Our results indicate that for such kernels, one can choose the centering measure for the Dirichlet process mixture model exactly as one would in the analogous simpler Dirichlet process model.
Keywords
Related Topics
Physical Sciences and Engineering
Mathematics
Statistics and Probability
Authors
Timothy Hanson, Jayaram Sethuraman, Ling Xu,