Posterior Simulation and Monte Carlo Methods
A beauty of the Bayesian approach is the principled nature of inference. There is a gold standard of how to proceed, and the basic principle is easily explained. However, the actual implementation often gives rise to many challenges. One of the challenges that remains an important research frontier of Bayesian inference is the problem of numerically evaluating the desired posterior summaries. In this chapter we review some related specific research problems.
KeywordsMarkov Chain Monte Carlo Importance Sampling Marginal Likelihood Markov Chain Monte Carlo Algorithm Inclusion Probability
Unable to display preview. Download preview PDF.