[R] names of model coefficients etc

[Prof Brian Ripley]

On Fri, 28 Sep 2001, David Firth wrote:

> I am curious about the reason for the following:
>
> >  xzw<-lm(y ~ x+cbind(z,w))
> >  names1<-names(xzw$coefficients)
> >  names1
> [1] "(Intercept)"  "x"            "cbind(z, w)z" "cbind(z, w)w"
> >  names2<-colnames(as.matrix(xzw$model))
> >  names2
> [1] "y"             "x"             "cbind(z, w).z" "cbind(z, w).w"
> >  names1==names2
> [1] FALSE  TRUE FALSE FALSE
>
> It would be helpful for some purposes if all components except the
> first here could be TRUE.  Presumably there is some good reason for
> the difference?  Can someone please enlighten me?

xzw$model is the model frame: xzw$coefficients refers to the model
matrix.  With factors in the model they are very different.

I guess you wanted a linear model fit with x=TRUE and xzw$x, as it

y <- rnorm(100)
x <- rnorm(100)
z <- rnorm(100)
w <- rnorm(100)
xzw<-lm(y ~ x+cbind(z,w), x = TRUE)
colnames(xzw$x)
[1] "(Intercept)"  "x"            "cbind(z, w)z" "cbind(z, w)w"

Further, be careful of calling xzw$coefficients rather than coef(xzw).
You will not always get the same thing with other fitting functions.


Brian

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._