MCMC Based Simple Significance Check for Smooth Terms

For each smooth term estimated with MCMC, the function computes 95 intervals and simply computes the fraction of the cases where the interval does not contain zero.

smooth_check(object, newdata = NULL, model = NULL, term = NULL, ...)

Arguments

object

A fitted model object which contains MCMC samples.

newdata

Optionally, use new data for computing the check.

model

Character, for which model should the check be computed?

term

Character, for which term should the check be computed?

...

Arguments passed to predict.bamlss.

Examples

if (FALSE) ## Simulate some data.
d <- GAMart()

## Model formula.
f <- list(
  num ~ s(x1) + s(x2) + s(x3),
  sigma ~ s(x1) + s(x2) + s(x3)
)

## Estimate model with MCMC.
b <- bamlss(f, data = d)
#> Error in eval(expr, envir, enclos): object 'd' not found

## Run the check, note that all variables
## for sigma should have no effect.
smooth_check(b)
#> Error in eval(expr, envir, enclos): object 'b' not found