Easy-to-Apply Results for Establishing Convergence of Markov Chains in Bayesian Analysis

Abstract

The Markov chain simulation method has become a powerful computational method in Bayesian analysis. The success of this method depends on the convergence of the Markov chain to its stationary distribution. We give two carefully stated theorems, whose conditions are easy to verify, that establish this convergence. We give versions of our conditions which are simpler to verify for the Markov chains that arise most commonly in Bayesian analysis.... Bayesian Poisson regression; Calculation of posterior distributions; Ergodic theorem; Markov chain simulation method.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Feb 01, 1993
Accession Number
ADA264015

Entities

People

  • Hani Doss
  • Jayaram Sethuraman
  • Krishna B. Athreya

Organizations

  • Florida State University

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Air Force
  • Case Studies
  • Computational Science
  • Convergence
  • Markov Chains
  • Mathematics
  • Military Research
  • Observation
  • Probability
  • Probability Distributions
  • Random Variables
  • Scientific Research
  • Simulations
  • Statistics
  • Stochastic Processes
  • Universities

Fields of Study

  • Mathematics

Readers

  • Mathematical Modeling and Probability Theory.
  • Statistical inference.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms