| toLoo {pcFactorStan} | R Documentation |
Compute approximate leave-one-out (LOO) cross-validation for Bayesian models using Pareto smoothed importance sampling (PSIS)
Description
You must use an ‘_ll’ model variation (see findModel).
Usage
toLoo(fit, ...)
Arguments
fit |
a |
... |
Additional options passed to |
Value
a loo object
See Also
Examples
palist <- letters[1:10]
df <- twoLevelGraph(palist, 300)
theta <- rnorm(length(palist))
names(theta) <- palist
df <- generateItem(df, theta, th=rep(0.5, 4))
df <- filterGraph(df)
df <- normalizeData(df)
dl <- prepCleanData(df)
dl$scale <- 1.5
m1 <- pcStan("unidim_ll", dl)
loo1 <- toLoo(m1, cores=1)
print(loo1)
[Package pcFactorStan version 1.5.4 Index]