Watanabe-Akaike Information Criterion (WAIC)

Function returning the Watanabe-Akaike Information Criterion (WAIC) of a fitted model object.

WAIC(object, ..., newdata = NULL)

Arguments

object

A fitted model object which contains MCMC samples.

...

Optionally more fitted model objects.

newdata

Optionally, use new data for computing the WAIC.

Value

A data frame containing the WAIC and estimated number of parameters.

References

Watanabe S. (2010). Asymptotic Equivalence of Bayes Cross Validation and Widely Applicable Information Criterion in Singular Learning Theory. The Journal of Machine Learning Research, 11, 3571--3594. https://jmlr.org/papers/v11/watanabe10a.html

Examples

if (FALSE) d <- GAMart()
b1 <- bamlss(num ~ s(x1), data = d)
#> Error in eval(expr, envir, enclos): object 'd' not found
b2 <- bamlss(num ~ s(x1) + s(x2), data = d)
#> Error in eval(expr, envir, enclos): object 'd' not found
WAIC(b1, b2)
#> Error in eval(expr, envir, enclos): object 'b1' not found