predict.SLik_j: Evaluate log-likelihood for given parameters

In Infusion: Inference Using Simulation

Evaluate log-likelihood for given parameters

Description

As the Title says. Implemented as a method of the predict generic, for objects created by the up-to-date workflow using gaussian mixture modelling of a joint distribution of parameters and statistics (hence the newdata argument, shared by many predict methods; but these newdata should be parameter values, not data).

Usage

## S3 method for class 'SLik_j'
predict(
  object, newdata, log = TRUE, which = "lik", 
  tstat = t(attr(object$logLs, "stat.obs")), 
  solve_t_chol_sigma_lists = object$clu_params$solve_t_chol_sigma_lists, 
  ...)

Arguments

object	an object of class SLik_j, as produced by infer_SLik_joint.
newdata	A matrix, whose rows each contain a full vector of the fitted parameters; or a single vector. If parameter names are not provided (as column names in the matrix case), then the vector is assumed to be ordered as object$colTypes$fittedPars.
log	Boolean: whether to return log-likelihood or likelihood.
which	"lik" or "safe". The latter protects against some artefacts of predictions beyond the regions of parameter space well sampled by the inference procedure.
tstat	The data (as projected summary statistics). Defaults to the data input in the inference procedure (i.e., the projected statistics used as stat.obs argument of infer_SLik_joint).
solve_t_chol_sigma_lists	For programming purposes. Do not change this argument.
...	For consistency with the generic. Currently ignored.

Value

Numeric: a single value, or a vector of (log-)likelihoods for different rows of the input newdata.

Examples

## see help("example_reftable")

Infusion documentation built on May 3, 2023, 5:10 p.m.