[R] Model comparison

Charles C. Berry cberry at tajo.ucsd.edu
Fri Mar 30 20:06:05 CEST 2007


On Fri, 30 Mar 2007, João Fadista wrote:

> Dear all,
>
> I would like to know if I can compare by a significance test 2 models 
> with different kind of parameters. Perhaps I am wrong but I think that 
> we can only compare 2 models if one is a sub model of the other.
>
>

The literature you seek is on 'non-nested' models. Here is a start:

Bradley Efron
Comparing non-nested linear models
J. of the Amer. Stat'l. Assn.
79:791-803,1984


Google is your friend. Try 'non-nested test' and 'non-nested model'
and you will get many hits.


[snip]
Charles C. Berry                        (858) 534-2098
                                          Dept of Family/Preventive Medicine
E mailto:cberry at tajo.ucsd.edu	         UC San Diego
http://biostat.ucsd.edu/~cberry/         La Jolla, San Diego 92093-0901



More information about the R-help mailing list