The usual way, I suppose: (1 - RSS (model with only a constant)/RSS(full model)).
However, it's not particularly meaningful to do this, because points are weighted differently depending on the model. This means that the usual interpretation as the proportion of unexplained variation is wrong; and none of the standard distributional properties hold either. Bottom line: Reconsider your request. Familiar linear model concepts and statistical results do NOT hold for robust regression, which is a kind of nonlinear regression, actually. Corrections and further clarifications by experts would be welcome. -- Bert On Thu, Sep 30, 2010 at 10:56 PM, Hock Ann Lim <lim...@yahoo.com> wrote: > May I know how to find the R-squared for robust regression model? > > Thank you. > > Hock Ann > > > > [[alternative HTML version deleted]] > > > ______________________________________________ > R-help@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. > > -- Bert Gunter Genentech Nonclinical Biostatistics ______________________________________________ R-help@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.
