Bayesian Variable Selection Using the Gibbs Sampler

Dipak K. Dey, Sujit Kumar Ghosh, Bani K. Mallick · CRC Press eBooks · 2000

In a Bayesian analysis of a generalized linear model, model uncertainty may be incorporated coherently by specifying prior probabilities for plausible models and calculating posterior probabilities using where m denotes the model, M is the set of all models under consideration, f(m) is the prior probability of model m. The observed data y contribute to the posterior model probabilities through f(y/m), the marginal likelihood calculated using f(y/m) =I f(y/m,/3m)f(f3m/m)df3m where f(f3m/m) is the conditional prior distribution of f3m, the model parameters for model m and f(y/m, f3m) is the likelihood of the data y under model m.

Cited by 50 works. doi:10.1201/9781482293456-25

Authors

References