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 <sumpr...@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@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 ______________________________________________ R-help@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.
