|
Activity Number:
|
506
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 10, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
IMS
|
| Abstract - #306558 |
|
Title:
|
Geometric Ergodicity of the Gibbs Sampler for the Probit Model
|
|
Author(s):
|
Vivekananda Roy*+
|
|
Companies:
|
University of Florida
|
|
Address:
|
Department of Statistics, Gainesville, FL, 32611-8545,
|
|
Keywords:
|
data augmentation ; convergence rate ; drift condition ; CLT
|
|
Abstract:
|
Let $\pi$ be the posterior distribution of the regression parameter $\beta$ that results when a probit regression model is combined with a uniform prior on $\beta$. Albert and Chib (1993)proposed a data augmentation algorithm for sampling from $\pi$. We show that the Markov Chain driving this algorithm is geometrically ergodic. This result has important practical ramifications concerning the calculation of standard errors.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2006 program |
