Extract log marginal likelihood from a model, saved as the first
element of a BayesMfp object.
getLogMargLik(x, design=getDesignMatrix(x), nObs = nrow(design), dim = ncol(design))
valid BayesMfp-Object of length 1 (otherwise only first
element recognized)
(centered) design matrix
number of observations
number of design matrix columns
Daniel Saban\'es Bov\'e
This function interfaces the C++ function logMargLik, and can
be used to compute the marginal likelihood of a model not saved in the
model list. But be careful to adjust the saved R^2 of the model, too,
and not only the powers! Therefore this function is internal only...
and is used e.g. in transformMfp.
getLogPrior