Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: Ordered probit postestimation


From   "Maarten Buis" <M.Buis@fsw.vu.nl>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Ordered probit postestimation
Date   Wed, 28 Feb 2007 14:04:40 +0100

-- Mckenzie, Lynda wrote:
> I am analysisng ordered categorical data with oprobit and would like to
> predict out of sample values.
>
> Eg: original variable variable1 has values 1, 2 or 3
> 
> Predicted variable v1, v2 and v3
> 
> When I type
> Mean v1 v2 v3, over(variable1) I get a table of mean values as follows.
> How do I interpret these mean values?
>
> V1  1  0.66
>     2 0.5
>     3 0.3

The output you show can't be the result of that command. Assuming that 
variable1 has three levels the output would be something like:

V1 1 0.66
   2 0.50
   3 0.15
V2 1 0.10
   2 0.10
   3 0.80
V3 1 0.24
   2 0.40
   3 0.05

This is interpreted as: For cases with a value of 1 on variable1 the 
average predicted probability to be belong to category 1 of the explained 
variable is 66%, 10% to belong to category 2, and 24% to category 3. For 
cases with a value of 2 on variable 1 the average predicted probability of 
belonging to category 1 of the explained variable is 50%, 10% for category 
2, and 40% to category 3. and so on. 


Have a look at -findit spost- for lots of tools that help with interpreting 
these kinds of models and at (Long and Freese 2006).

Hope this helps,
Maarten

J. Scott Long and Jeremy Freese (2006) "Regression Models for Categorical 
Dependent Variables Using Stata", Stata Press.


-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology 
Vrije Universiteit Amsterdam 
Boelelaan 1081 
1081 HV Amsterdam 
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434 

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


    

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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   |   What's new   |   Site index