[R] Regression: standardized coefficients & CI

Bert Gunter gunter.berton at gene.com
Wed Nov 21 22:10:21 CET 2012


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 <torvon at 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 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




More information about the R-help mailing list