Merge-split proposals for conjugate "Chinese Restaurant Process" (CRP) mixture models using sequentially-allocated elements. Allocation is performed with weights derived from a previously-calculated pairwise similarity matrix.
psmMergeSplit_base(
partition,
psm,
logPosteriorPredictiveDensity = function(i, subset) 0,
mass = 1,
discount = 0,
nUpdates = 1L,
selectionWeights = NULL
)A numeric vector of cluster labels representing the current partition.
A matrix of previously-calculated pairwise similarity probabilities for each pair of data indices.
A function taking an index \(i\) (as a numeric vector of length one) and a subset of integers \(subset\), and returning the natural logarithm of \(p( y_i | y_subset )\), i.e., that item's contribution to the log integrated likelihood given a subset of the other items. The default value "turns off" the likelihood, resulting in prior simulation (rather than posterior simulation).
A specification of the mass (concentration) parameter in the CRP
prior. Must be greater than the -discount argument.
A numeric value on the interval [0,1) corresponding to the discount parameter in the two-parameter CRP prior.
An integer giving the number of merge-split proposals before returning. This has the effect of thinning the Markov chain.
A matrix or data frame whose first two columns are the unique pairs of data indices, along with a column of weights representing how likely each pair is to be selected at the beginning of each merge-split update.
A numeric vector giving the updated partition encoded using cluster labels.
The acceptance rate
of the Metropolis-Hastings proposals, i.e. the number of accepted proposals
divided by nUpdates.