Re: S-PLUS Vs some other softwares

[Martin Maechler <maechler@stat.math.ethz.ch>]

>>>>> "BertG" == Gunter, Bert <bert_gunter@merck.com>
>>>>>     on Tue, 2 Mar 2004 08:57:01 -0500 writes:

    BertG> As Andy said, lmList will probably do what you want,
    BertG> but as it uses lm(), it may also take a while. If you
    BertG> wish to do it "by hand" yourself, try by() [which is
    BertG> a wrapper for tapply()] and use lsfit instead of
    BertG> lm. As others have said, lm may incur a lot of
    BertG> overhead. The code would be something like
 
    .................

I think it should be lm.fit() , not lsfit().
This will use the numerical engine below lm() without
the overhead of formula -> model.frame -> model.matrix.

Martin Maechler <maechler@stat.math.ethz.ch>    http://stat.ethz.ch/~maechler/
Seminar fuer Statistik, ETH-Zentrum  LEO C16    Leonhardstr. 27
ETH (Federal Inst. Technology)  8092 Zurich     SWITZERLAND
phone: x-41-1-632-3408          fax: ...-1228                   <><