Thanks to Kit Baum, a revised version of the -intcens- package is now
available
for download from SSC. In Stata, use the -ssc- command to do this.
The program fits various parametric models for non-negative outcomes to
data which can be interval-censored, or point data, left- or
right-censored.
The distributions are all those fitted by -streg-, plus the
two-parameter gamma, inverse Gaussian (which is the time to reach a
certain distance from the origin for a Weiner process) and an extension
of the inverse Gaussian which is the time to reach a certain point for a
Weiner process with random drift.
The package also contains a do file and a SAS program to check the
results, for people who have SAS.
There was a bug in the code for the Gompertz and generalized gamma
distributions which has been fixed.
In the likelihood evaluator, in the code for these two distributions, I
referred to a parameter as if it was a scalar when in fact with maximum
likelihood method -lf-, all the parameters are temporary variables. The
code worked by accident as long as the first observation in the dataset
was in the estimation sample, but issued an error message otherwise.
When the program did produce results, then they were correct.
Jamie Griffin.
*
* 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/
