[R] robust regression

David Winsemius dwinsemius at comcast.net
Thu Jan 22 17:53:17 CET 2009


On Jan 22, 2009, at 11:06 AM, Georg Ehret wrote:

> Dear Miss R,
>   I have a large dataset that is skewed and would like to calculate
> confidence intervals for my regression line. Unfortunately I have  
> trouble

> finding commands for rubust regression and in particular for the  
> confidence
> intervals... Can you possibly give a hint?

Such difficulty is inexplicable.

Search with Google for "robust regression" and add r-project to your  
search string for narrowing.
(First hit is John Fox's 4 page addendum to CAR with worked examples.  
Many hundreds of hits but with high chance that the count is inflated.  
Further addition of "confidence intervals" still produces hundreds of  
results.)

Search r-site-search for "robust regression"
http://search.r-project.org/nmz.html
(roughly 300 hits)

See also the CRAN TAsk View on Robust Statistical Methods:
http://cran.r-project.org/web/views/Robust.html

(I have had very satisfactory results with Koenker's quantreg package  
functions, but you have provided insufficient description of a) the  
data, b) the domain of analysis, or c) your goals to allow any comment  
on the need or applicability of methods to your tasks.)

-- 
David Winsemius

>
> Thank you and best regards, Gerog.
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> 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.




More information about the R-help mailing list