Hello I have a data set with outlier and it is not normally distributed. I would instead like to use a more robust distribution like t-distribution. My question is if the coefficients of the regression are different from zero, but assuming a t-distribution. Could someone hint me what package to use or.... Thanks in advance Rosario
did you try
RSiteSearch("t-regression")
?
That seems to give some usefull hits.
On Mon, Feb 21, 2011 at 7:42 AM, Rosario Garcia Gil
<M.Rosario.Garcia at slu.se> wrote:> Hello
>
> I have a data set with outlier and it is not normally distributed. I would
instead like to use a more robust distribution like t-distribution.
>
> My question is if the coefficients of the regression are different from
zero, but assuming a t-distribution.
>
> Could someone hint me what package to use or....
>
> Thanks in advance
> Rosario
> ______________________________________________
> 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.
>
On 2011-02-21 02:42, Rosario Garcia Gil wrote:> Hello > > I have a data set with outlier and it is not normally distributed. I would instead like to use a more robust distribution like t-distribution. > > My question is if the coefficients of the regression are different from zero, but assuming a t-distribution. > > Could someone hint me what package to use or....Have you considered using a robust method? There's rlm() in MASS, lmrob() in robustbase, etc. Peter Ehlers> > Thanks in advance > Rosario > ______________________________________________ > 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.