Dear Pushpa Methekar,
You can use which.max(), as in
> library(car)
> mod <- lm(prestige~ income + education, data=Duncan)
> which.max(abs(rstudent(mod)))
minister
6
In fact, this is exactly what outlierTest() does, as you can see by looking at
car:::outlierTest.lm
Beyond that, I'd encourage you to look more carefully at potential outliers,
including at the distribution of the studentized residuals, rather than
adopting a simple mechanical rule for removing outliers.
I hope this helps,
John
------------------------------------------------
John Fox, Professor
McMaster University
Hamilton, Ontario, Canada
http://socserv.mcmaster.ca/jfox/
On Mon, 5 Jan 2015 12:20:15 +0000
"Methekar, Pushpa (GE Transportation, Non-GE)" <pushpa.methe...@ge.com> wrote:
>
>
> Hi, this is my function to find rstudent of model which will give me outlier
> But I wonder if I could find out the exact no. of outlier in data set.
> Like outlierTest() does from car package .
>
>
> rm.outliers = function(dataset,model){
> dataset$rstudent = rstudent(model)
> for(i in 1:length(dataset$rstudent)){
> if(dataset$rstudent[i]>=-3 & dataset$rstudent[i]<=3)
> {print(dataset$rstudent[i])}
> else { print("this is an outlier")
> library(car)
> print(outlierTest(model,n.max=1, order=TRUE) ) }
> i<-i+1
> }
> }
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
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.
