[R] SSEc and SSEr

[Bert Gunter]

Sadly, your commonly held belief is wrong (imho) -- p
values/statistical significance are not a legitimate decision criteria
for model "appropriateness," especially scientific appropriateness.
That requires more careful consideration of a relevant "utility
function" (to use Frank Harrell's phrase), effect sizes, power, etc.,
a more detailed discussion of which belongs elsewhere, not here, as
this has nothing to do with R.

For that reason, anyone with a contrary opinion on this -- there may
be many who disagree -- should reply personally offlist.

Cheers,
Bert

On Thu, Jul 26, 2012 at 9:14 AM, suman kumar <sumprain at gmail.com> wrote:
> You can make different lm objects by adding all predictors and compare them
> with anova(lm1,lm2,lm3...).  See if p value is not significant, the more
> complex model is not appropriate.
> Dr Suman Kumar
>
>
>
> --
> View this message in context: http://r.789695.n4.nabble.com/SSEc-and-SSEr-tp4637855p4637963.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.



-- 

Bert Gunter
Genentech Nonclinical Biostatistics

Internal Contact Info:
Phone: 467-7374
Website:
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm