[R] enhanced Question to stand. Beta

Thomas Lumley tlumley at u.washington.edu
Tue Mar 5 17:21:36 CET 2002


On Tue, 5 Mar 2002, vito muggeo wrote:

> >a question that connect to the question of Frederik Karlsons >about 'how
> >to stand. betas'
> >With the stand. betas i can compare the influence of the >different
> >explaning variables. What do i with the betas of factors? I can't >use
> >the solution of JohnFox, because there is no sd of an factor. >How can i
> >compare the influence of the factor with the influence of the >numeric
> >variables?
>
>
> Use the t-values from summary(lm(....))!
> Are not the standardized regression coefficients the "t values"?
>

Not at all. The t-values are coefficients divided by their standard
errors. Standardised coefficients are divided by the standard deviation of
the predictor.

There isn't really a way to do this for factors.

Consider the much simpler case of binary data. You could use the standard
deviation of the variable, but there might be just as strong a case for
considering the variable as a discretisation of a Normal and working out
some sort of latent variable model.  These approaches would typically give
quite different results, suggesting that the question doesn't have a
unique answer.


	-thomas



-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list