Re: [R] regression with more than one observation per x value

From: Liaw, Andy <andy_liaw_at_merck.com>
Date: Wed 17 Aug 2005 - 00:35:00 EST


Sounds like you're looking for something like pure.error.anova in the `alr3' package on CRAN...

Andy

> From: Christoph Scherber
>
> Dear R users,
>
> How can I do a regression analysis in R where there is more than one
> observation per x value? I tried the example in Sokal&Rohlf
> (3rd edn.,
> 1995), page 476 ff., but I somehow couldnīt find a way to
> partition the
> sums of squares into "linear regression", "deviations from
> regression",
> and within-groups SS.
>
> I tried
>
> model1<-lm(y~as.numeric(x)+as.factor(x) #with treatment contrasts
>
>
> but I am sure thereīs a better way around it. I would be very
> happy if
> anyone could give me some suggestions on this.
>
> Best regards
> Christoph.
>
> ______________________________________________
> R-help@stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
> http://www.R-project.org/posting-guide.html
>
>
>



R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Wed Aug 17 00:42:03 2005

This archive was generated by hypermail 2.1.8 : Sun 23 Oct 2005 - 15:23:34 EST