[RsR] Wanted: predict.lmrob

Locher René (lore) lore at zhaw.ch
Thu Sep 13 19:00:18 CEST 2007


Dear Matias

predict.lmrob() seems not to exist and predict.lm() gives an incorrect result. Is there another convenient trick to predict other data than the fitted? Below you 'll find a short example.


library(robustbase)
xy <- data.frame(x=1:10, y=(1:10)+rnorm(10))

res <- lmrob(y~x, xy)

fitted(res)
## correct

predict(res,newdata=data.frame(x=10:20))
## Error in predict(res, newdata = data.frame(x = 10:20)) : 
##	no applicable method for "predict"

predict.lm(res,newdata=data.frame(x=10:20))
## does not work properly!

sessionInfo()
## R version 2.5.1 (2007-06-27) 
## i386-pc-mingw32 
## 
## locale:
## LC_COLLATE=German_Switzerland.1252;LC_CTYPE=German_Switzerland.1252;LC_MONETARY=German_Switzerland.1252;LC_NUMERIC=C;LC_TIME=German_Switzerland.1252
## 
## attached base packages:
## [1] "stats"     "graphics"  "grDevices" "utils"     "datasets"  "methods"  
## [7] "base"     
## 
## other attached packages:
## robustbase 
##    "0.2-8" 


Kind regards and thanks in advance

René

René Locher                                       E-Mail: rene.locher at zhaw.ch
Institut für Datenanalyse und Prozessdesign       Tel:    +41 58 934 7810
Zürcher Hochschule für angewandte Wissenschaften  Fax:    +41 58 935 7810
Rosenstrasse 3                   
Postfach
CH-8400 Winterthur                                http://www.idp.zhaw.ch



More information about the R-SIG-Robust mailing list