This function calculates the log-likelihood of a tsbart model fit, and optionally the WAIC.

checkFit(y, mcmcdraws, probit, doWaic, sig=NULL, yobs=NULL)

Arguments

y

A vector with continuous response values. Required, defaults to NULL.

mcmcdraws

A matrix containing the MCMC draws for the tsbart fit. Rows are MCMC iterations, columns are observations. Number of columns must match length of y. Required, defaults to NULL.

probit

Boolean indicator for whether model is probit.

doWaic

Boolean indicator for whether to calculate WAIC in addition to log-likelihood.

sig

A vector with sigma MCMC draws. Required if probit=FALSE, defaults to NULL. Must be NULL if probit=TRUE.

yobs

A vector of 0/1 indicators for each observation. Required if probit=TRUE, defaults to NULL.

Value

A list containing the following components:

waic

The waic value for the model fit.

ll_mcmc

The log-likelihood for all mcmc draws.

Examples

if (FALSE) { # Function call example is as follows. Let fit be the output of tsbart(). ## In-sample example: checkFit(y=y, mcmcdraws = fit[["mcmcdraws"]], probit=FALSE, doWaic=TRUE, sig = fit[["sigma"]], yobs = NULL) ## Out-of-sample example: checkFit(y=y_pred, mcmcdraws = fit[["mcmcdraws_oos"]], probit=FALSE, doWaic=TRUE, sig = fit[["sigma"]], yobs = NULL) }