predict.cubinf: Prediction methods for objects of class "cubinf"

Description

Predictions provided by a model fit when method is "cubinf".

Usage

# S3 method for cubinf
predict(object, newdata, type = c("link", "response", "terms"), 
   se.fit = FALSE, terms = labels(object$terms), ...)

Value

The value returned depends on type.

Arguments

object: An object of class "cubinf" for which predictions are desired.
newdata: Specify the explanatory variables to used.
type: The prediction type.
se.fit: Logical to specify if standard errors are returned or not.
terms: The terms in newdata.
...: Additional arguments affecting the predictions produced.

References

Marazzi, A. (1993). Algorithms, Routines, and S-functions for robust Statistics. Chapman and Hall, New York.

Kuensch, H.R., Stefanski L.A., Carroll R.J. (1989). Conditionally unbiased bounded-influence estimation in general regression models, with application to generalized linear models. Journal of the American Statistical Association, 84, 460-466.

Examples

Run this code

 library(robcbi)
 data(Finney)
 Vol <- Finney$Vol; Rate <- Finney$Rate; Resp <- Finney$Resp
 df <- data.frame(lVol = log(Vol), lRate = log(Rate), Resp = Resp)
 z.cub <- glm(Resp~lVol+lRate,family=binomial,data=df,method="cubinf",ufact=3.2)
 set.seed(123)
 rVol  <- runif(20,0.4,3.7); rRate <- runif(20,0.3,3.75)
 newdat <- data.frame(lVol=log(rVol),lRate=log(rRate))
 predict(z.cub, newdat, type="response")