Bert, If I am not mistaken the CI provided by confint(fit) are for the unstandardized beta weights, not the standardized. Although I found a tutorial for getting the standardized beta weights ( http://polisci.msu.edu/jacoby/msu/pls802/handouts/stdized/Stdized%20Coeffs%20in%20R,%20Handout.pdf), I still don't know how to get the CIs for these values seeing that they are manually computed.
Thanks On 21 November 2012 19:10, Bert Gunter <gunter.ber...@gene.com> wrote: > ?confint > > -- Bert > > On Wed, Nov 21, 2012 at 3:55 PM, Torvon <tor...@gmail.com> wrote: > > Bert, > > > > Please excuse me, and let me rephrase: > > > > How do I obtain the confidence intervals of the _standardized_ beta > weights > > for predictors in a linear regression in R? > > > > Thank you. > > Torvon > > > > > > On 21 November 2012 16:10, Bert Gunter <gunter.ber...@gene.com> wrote: > >> > >> 1. This is a statistics, not an R, question. Post on a statistics > >> list, like stats.stackexchange.com > >> > >> Also... > >> > >> On Wed, Nov 21, 2012 at 12:39 PM, Torvon <tor...@gmail.com> wrote: > >> > I run 9 WLS regressions in R, with 7 predictors each. > >> > > >> > What I want to do now is compare: > >> > (1) The strength of predictors within each model (assuming all > >> > predictors > >> > are significant). That is, I want to say whether x1 is stronger than > x2, > >> > and also say whether it is **significantly stronger.** > >> > >> -- I have no idea what this means, though perhaps it is defined > >> somewhere and in some way that I am not familiar with. When you post > >> to a stats list, I suggest you provide a reference so the folks there > >> know what you mean by this. > >> > >> -- Bert > >> > >> I compare strength by > >> > simply comparing standardized beta weights, correct? How do I compare > if > >> > one predictor is significantly stronger than the others? I thought > about > >> > comparing confidence intervals, but if I understand correctly the > >> > confidence intervals are calculated from the unstandardized beta > >> > weights, > >> > which in this case would not help me, correct? > >> > (2) The strength of the same predictor over different models. I want > to > >> > say > >> > whether x1 affects y1 - y9 equally strong or not. How would I do this? > >> > > >> > I hope that I provided all information that is needed. > >> > Thank you > >> > T > >> > > >> > [[alternative HTML version deleted]] > >> > > >> > ______________________________________________ > >> > 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 > > > > > > > > -- > > 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 > [[alternative HTML version deleted]] ______________________________________________ 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.
