[R] Standardised coeff. & model fit

Tord Snall tord.snall at ebc.uu.se
Tue Jul 18 18:35:23 CEST 2000

Dear all,

I have developed regression models using stepAIC() but since the aim of my
study is to show the relative importance of different parameters I want the
standardised partial regression coefficients, their standard deviations and
confidence intervals. The models are based on data from binomial or Poisson
distributions. Can I get those estimates from R or do I have to rerun the
analysis using normalised data? 

If the estimates are not available in R I plan to normalise by subtracting
each value from the mean of that variable, thereafter dividing with the
standard deviation of that variable. I assume there is no problem in
normalising binary independent variables.

I plan to present each model (the linear component) in a histogram where
the height and direction of the bars show the standardised coefficients and
their sign, respectively. Error bars will show the 95%-confidence interval.

I also would like to know what value is normally presented for summarising
fit when the modelling is based on AIC. Is it model AIC or is it mean
deviance? In ecological literature I have seen "% explained deviance"
assuming it is (Null Dev. - Resid. Dev.) / Null Dev. but I am not really
satisfied with this.

I am using R 1.1.0 on Windows NT.

Thanks in advance for any comments or help!

Tord Snäll

******New e-mail address: Tord.Snall at ebc.uu.se*******

Tord Snäll, PhD student
Avd. f växtekologi, Evolutionsbiologiskt centrum, Uppsala universitet
Dept. of Plant Ecology, Evolutionary Biology Centre, Uppsala University
Villavägen 14			
SE-752 36 Uppsala, Sweden
Tel: 018-471 28 82 (int +46 18 471 28 82) (work)
Tel: 018-25 71 33 (int +46 18 25 71 33) (home)
Fax: 018-55 34 19 (int +46 18 55 34 19) (work)
E-mail: Tord.Snall at ebc.uu.se
Homepage: http://www.vaxtbio.uu.se/resfold/tord_s.htm
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