Identifying groups of determinants in Bayesian model averaging using Dirichlet process clustering

Model uncertainty is a pervasive problem in regression applications. Bayesian model averaging (BMA) takes into account and identifies robust determinants. However, it requires the specification of suitable priors. Mixture priors are appealing because they explicitly for different groups covariates as Specific Dirichlet process clustering (DPC) proposed; their correspondence to binomial prior derived methods perform BMA analysis including DPC postprocessing procedure identify determinants outlined. The application these demonstrated simulation exercise an empirical cross-country economic growth data. performed using Markov chain Monte Carlo composition sampler obtain samples from posterior specifications. Results compared with those obtained under beta-binomial collinearity-adjusted dilution prior.