[R] Straight

Zsombor Cseres-Gergely z.cseres-gergely at ucl.ac.uk
Wed Nov 1 18:45:09 CET 2000

```On Wed, Nov 01, 2000 at 08:23:36AM +0000, Prof Brian Ripley wrote:

> What do you want there: estimation under restrictions or tests of linear
> combinations?  The latter is a simple R programming exercise. I don't know

Sorry for bad wording, I meant the latter. And this is why I did not ask this
before: easy, _provided_ we have the covariance matrix of the estimators. Now
I have the answer also for this, so it is really straight. (I still do not
know, how summarize() gets it, since in a lm.model, I have not found the
underlying data for it [or overlooked]. The reason I did not do a X'X was that
I believe that this is expensive. I have not tried, but my previous
experiences (again: Stata) tells me this. Or am I wrong, and this is cheap in R?

> how to do the first easily (nor do I see why I would need to: normally one
> can re-parametrize and test nested models via anova),.

Sure, this is not an issue.

> MASS has confidence intervals (generic function confint), but again,
> a simple R programming exercise for linear models.

Same appies.

Thank you,

Zsombor
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

```