Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: extracting test statistics and p values for cut points in oprobit models


From   "maartenbuis" <[email protected]>
To   [email protected]
Subject   Re: st: extracting test statistics and p values for cut points in oprobit models
Date   Fri, 01 Jul 2005 16:07:25 -0000

---  Nathan Barleen <barl0050@u...> wrote:
> I have estimated several ordered probit models but there has been 
> some question about the significance of the ancillary parameters 
> (cut points).  The test statistics and pvalues of the estimated 
cut 
> points are not displayed.  Is there any way to extract these 
values 
> from stata?

Hi Nathan,

e(V) gives you the variance covariance matrix for all parameters 
including the cut points so you can test any null hypothesis you 
want. However, testing whether the cuts are equal to zero is hardly 
meaningfull. The relative distances between categories are derived 
from the data, but the location (where is point zero) and scale are 
unidentified and are arbitrarily fixed at some point. A meaningfull 
test would be to see if categories could be joined together, i.e. if 
the cuts are significantly different. You can do this by hand by 
looking up the appropriate variances and covariances in e(V) or use 
the -test- command. 

Hope this helps,
Maarten



*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index