Line for the server...
Also note this recent discussion:
http://www.stata.com/statalist/archive/2008-10/msg00251.html
HTH
Martin
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Nick Cox
Sent: Wednesday, January 14, 2009 1:37 PM
To: [email protected]
Subject: st: RE: Re: categorical variables
That is correct. However, note also that Stata treats predictors as
categorical in -anova- unless you specify otherwise. Using -xi:- as
framework is perhaps the most common way to specify categorical
predictors.
Nick
[email protected]
Kit Baum
(3) Stata has no concept of 'categorical variables'. Numeric variables
may be stored as various integer data types, which would correspond to
categorical variables, as well as various real data types. As other
posters have suggested, using something like -codebook- or -inspect-
may be helpful. Also note that missing values are the largest positive
numbers, so beware of statements like 'if response > 4', as that will
include observations with a missing response.
On Jan 14, 2009, at 02:33 , Iant wrote:
> I am very new to STATA and would like to ensure that while running a
> model, STATA understands that a certain variable is categorical
> (actually binary response, either yes or no). It seems from the
> response post that as long as the values I have entered for the
> variable
> are binary, that STATA will understand the variable as categorical.
*
* 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/
*
* 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/
