Plots the posterior model probabilites based on 1) marginal likelihoods and
2) MCMC frequencies for the best models in a 'bma' object and details the
sampler's convergence by their correlation
A call to bms with a MCMC sampler (e.g.
bms(datafls,mcmc="bd",nmodel=100) uses a Metropolis-Hastings
algorithm to sample through the model space: the frequency of how often
models are drawn converges to the distribution of their posterior marginal
likelihoods. While sampling, each 'bma' object stores the best models
encountered by its sampling chain with their marginal likelihood and their
MCMC frequencies. plotConv compares the MCMC frequencies to
marginal likelihoods, and thus visualizes how well the sampler has
converged.
See Also
pmp.bma for posterior model probabilites based on the
two concepts, bms for creating objects of class 'bma'