calc_relative_lik_matrix: Calculate matrix of relative likelihoods.
In mashr: Multivariate Adaptive Shrinkage

calc_relative_lik_matrix

R Documentation

Calculate matrix of relative likelihoods.

Computes matrix of relative likelihoods for each of J rows of Bhat for each of P prior covariances.

This is an internal (non-exported) function. This help page provides additional documentation mainly intended for developers and expert users.

calc_relative_lik_matrix(data, Ulist, algorithm.version = c("Rcpp", "R"))

`data`	A `mash` data object; e.g., created by `mash_set_data`.
`Ulist`	List containing the prior covariance matrices.
`algorithm.version`	indicates R or Rcpp

The return value is a list containing the following components:

`lik_matrix`	J x P matrix containing likelihoods p(bhat[j] + Ulist[p], V), but normalized so that the largest entry in each row is 1.
`lfactors`	Vector which will recover the original likelihoods; for example, `lfactors[i] + log(lik_matrix[i,])` yields the log-likelihoods corresponding to row i of the Bhat data matrix.