[R] finding best fitting model

[Dieter Menne]

Benedikt Niesterok wrote:
> 
> Is there a function in r to find the best fitting model for a set of data?
> I would like to know if my data are related exponentially,linearly or if
> there is a logarithmic correlation between my x and y values.
> 

There is no one-stop method to do this. I would suggest that you try the
alternatives by doing the transformation (implicit in lm or explicit on
data) and checking the residuals of the fit.

Check documentation of plot.lm for an example or get the chapter in MASS on
the subject.

Dieter

-- 
View this message in context: http://www.nabble.com/finding-best-fitting-model-tp22852224p22864012.html
Sent from the R help mailing list archive at Nabble.com.