R help - May 2011 - errorest of library iprde: specifiying parameters for regularized DA

Hi dear R users,

I am using the function errorest from the libbrary ipred to estimate the 10 fold
and bootstrap classification error rate using regularized discriminat function
analysis (functiopn rda from library klaR). Is it possible to set my own values
for gamma and lambda?
Any help will be highly appreciated!

Best wishes, 

Felix

library(klaR)
library(ipred)

# force predict to return class labels only
mypredict.lda <- function(object, newdata) predict(object, newdata =
newdata)$class

# 10-fold cv of RDA for Iris data
errorest(Species ~ ., data=iris, model=rda, estimator = "cv", predict=
mypredict.lda)
--

On 27.05.2011 11:55, Felix Bach wrote:
> Hi dear R users,
>
> I am using the function errorest from the libbrary ipred to estimate the 10
fold and bootstrap classification error rate using regularized discriminat
function analysis (functiopn rda from library klaR). Is it possible to set my
own values for gamma and lambda?
> Any help will be highly appreciated!
>
> Best wishes,
>
> Felix
>
> library(klaR)
> library(ipred)
>
> # force predict to return class labels only
> mypredict.lda<- function(object, newdata) predict(object, newdata =
newdata)$class
>
> # 10-fold cv of RDA for Iris data
> errorest(Species ~ ., data=iris, model=rda, estimator = "cv",
predict= mypredict.lda)

Yes, via ... as in:

errorest(Species ~ ., data=iris, model=rda, estimator = "cv",
  predict= mypredict.lda, gamma=0, lambda=0)

Uwe Ligges


> --
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.