[R] advice on semi-serious attempt to extend summary

ivo welch ivowel at gmail.com
Mon Jan 1 22:03:50 CET 2007

Dear R wizards:

I am trying (finally) to build a function that might be useful to
others.  In particular, I want to create a summary.lme (extended lm)
method that [a] adds normalized coefficients and [b] white
heteroskedasticity adjusted se's and T's.  I believe I already know
how to do the programming to do these two, at least in simple
unweighted cases.  Now my challenges are just [1] to trap weird cases
(e.g., hccm dies because the standard errors cannot be computed), and
[2] to follow "proper R rules and regulations."

I started my experiments by copying summary.lm() to summary.lme.  But
there is some magic that I do not understand.  Apparently, the
  class(ans) <- "summary.lm"
signals to R that it should not produce an unlisted summary of the
components of ans, but that it should print something that is nicely
  example:  y=rnorm(5); x=rnorm(5); m=lm(y~x);  summary.lm(m);
does exactly this nice output.  alas, just replacing the name at the end with
    class(ans) <- "summary.lme"
and typing summary.lme(m), having renamed summary.lm to summary, then
loses the nice printout.

so, it seems to me that somewhere R knows that a "summary.lm" object
gets special printing.  if I have not misunderstood this, then where
and how does this magic happen?

advice appreciated.



More information about the R-help mailing list