st: EM Algorithm for left censored data

[Jörg Hartmann <joerg.hartmann@sowi.uni-goettingen.de>]

I want to use event history analysis with data that contain a large share of 
left-censored episodes. There have been a few research papers recently 
that apply Expectation-Maximization (EM) algorithms to obtain Maximum-
Likelihood parameters, for example Kapetanakis et al. 2012: A semi-
Markov model stroke with piecewise-constant hazards in the presence of 
left, right and interval censoring. Statistics of Medicine.

Kapetanakis model looks good for my purposes, however, I am stuck with 
the following problems:

(1) there seems to be no Stata package providing Kapetanakis model, at 
least none that I am aware of

(2) in the case that I have to adopt existing EM algorithms in Stata 
(something I am not sure how to do yet), I don't find any EM algorithm 
package

Is there something can help me is applying this approach to left censored 
data? Or any other approaches that I am not aware of?

kind regards,
Jörg Hartmann
-- 


M.A. Jörg Hartmann
Wissenschaftliche Mitarbeiter
Lehrstuhl Prof. Dr. Karin Kurz 

Institut für Soziologie 
Georg-August-Universität Göttingen 
Platz der Göttinger Sieben 3 
37073 Göttingen 
Raum: Oeconomicum 0.105 

Tel. +49 (0) 551-39-7198 

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/