Closed pph23 closed 2 weeks ago
The predict()
method for gamlss
objects is provided in both the current CRAN and the current development version on R-universe.
library("gamlss")
m <- gamlss(dist ~ pb(speed), data = cars)
## GAMLSS-RS iteration 1: Global Deviance = 411.3478
## GAMLSS-RS iteration 2: Global Deviance = 411.3478
predict(m, newdata = data.frame(speed = 20))
## [1] 60.68918
However, the underlying function predict.gamlss
is not fully exported and thus you cannot access it directly. It is available in the package, though, and you should call it through the generic predict()
as above. If you need to access the source code, you can do so by using the fully-qualified name with the :::
notation:
predict.gamlss
## Error: object 'predict.gamlss' not found
gamlss:::predict.gamlss
## function (object, what = c("mu", "sigma", "nu", "tau"), parameter = NULL,
## newdata = NULL, type = c("link", "response", "terms"), terms = NULL,
## se.fit = FALSE, data = NULL, ...)
## {
## concat <- function(..., names = NULL) {
## tmp <- list(...)
## [...]
'predict.gamlss' function is not working/not available in the current version.