--- 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/