Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: AW: [Fwd: st: AW: GLM family and link (default)]
From
Maarten buis <[email protected]>
To
[email protected]
Subject
Re: AW: [Fwd: st: AW: GLM family and link (default)]
Date
Mon, 14 Jun 2010 13:37:59 +0000 (GMT)
--- On Mon, 14/6/10, Martin Weiss wrote:
> It is quite easy to replicate "mmolina"´s problem, btw:
>
>
> *************
> sysuse auto, clear
> set seed 43254
>
> prob foreign weight length trunk turn /*
> */ displacement headroom price mpg rep78 /*
> */ if runiform()<.3
>
> glm foreign weight length trunk turn /*
> */ displacement headroom price mpg rep78 /*
> */ if e(sample), family(binomial 1) /*
> */ link(probit)
> *************
That is actually not quite true, the problem you
are creating is similar but not the same. In your
case the combination of 9 variables together could
perfectly predict your outcome variable, while in
mmolina's case there were two variables, each of
which could perfectly predict the outcome on their
own.
Anyhow, we agree mmolina's model contains way too
many variables given her sample size.
Hope this helps,
Maarten
--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://www.maartenbuis.nl
--------------------------
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/