Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Fwd: st: AW: GLM family and link (default)]


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: [Fwd: st: AW: GLM family and link (default)]
Date   Mon, 14 Jun 2010 13:08:53 +0000 (GMT)

--- On Mon, 14/6/10, mmolina@uniroma3.it wrote:
> They are not 33 observations but these are the remaining...

It doesn't matter how large your dataset is, all that counts
is how many observations are used in your estimation, and
that is 33 (and only 22 in your probit model), which is a 
major problem when you want to estimate 13 parameters. Anyhow, 
the linear probability model is not the most obvious solution 
to your problem: 

First, simplify your model by using much less variables, 
I'd say an absolute maximum of 3 variables (10 obs per 
variable). 

Second, use -exlogisitc- if you want to retain the variables 
that perfectly predict your outcome, which as I stated before
was the reason that your -probit- model was mis-behaving.

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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index