[R] regression fit

[David Winsemius]

On Nov 25, 2011, at 2:56 PM, magushi wrote:

> Hi every one.
> I want to analyze my data. The  y (density) is very skewed to the  
> right.
> this is my tentative fit<-density~factor(Site)+factor(Species)+ depth.
> which model to apply between glm and lm, for sure I can not use  
> lm( This is
> non linear data)

Until you have looked more carefully at your tentative fit you are  
jumping to unsupported assumptions. It is a common misconception that  
a skewed distribution of the dependent variable represents a problem  
for linear methods. It may be true but you will not know until you  
look at the distribution of the residuals of a linear fit. So far we  
see not evidence that you have done so. If you showed us plots of  
density ~ depth within categories of Site and Species then there might  
be a basis for discussion of alternative links or transformations.

This wouldn't be homework by any chance, would it?


>
> --
> View this message in context: http://r.789695.n4.nabble.com/regression-fit-tp4108686p4108686.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> 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.

David Winsemius, MD
West Hartford, CT