|
Activity Number:
|
506
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 3, 2016 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
|
Abstract #319286
|
|
|
Title:
|
Bayesian Computation in Dirichlet Process Mixture Models
|
|
Author(s):
|
Erina Paul* and Sanjib Basu
|
|
Companies:
|
and Northern Illinois University
|
|
Keywords:
|
Approximate Bayesian Computation ;
Dirichlet process ;
non-conjugate prior
|
|
Abstract:
|
Dirichlet process mixture (DPM) provides a flexible and popular model for nonparametric Bayesian analysis with a substantial literature on Markov chain sampling and other computation methods for conjugate and non-conjugate DPM models. In recent years, approximate Bayesian computation methods have been successfully applied in Bayesian analysis of complex models where the likelihood function is analytically unavailable or computationally intractable. These methods provide samples from a distribution which is close to the target posterior distribution of interest. We propose an efficient computation method for DPM using approximate Bayesian computation and illustrate the performance of the proposed method in real dataset.
|
Authors who are presenting talks have a * after their name.
