Hello Stata Listers,
I am writing an ML program to simultaneously estimate parameters for a sequential decision problem in a binary logit setting (two rounds of decisions).
In round 2 the independent variable is a function of the parameter estimate from round 1.
I can do this quite easily sequentially by using logit to estimate round 1 parameters and then do some intervening calculations to create the independent variable for round 2, and then apply logit again.
For greater efficiency, and to relax assumptions about the decision process, I would like to estimate the parameters simultaneously. Since the dependence of round 2's contribution depends on round 1 decisions this violates the linear form and so I am attempting to use d0 method.
The problem I encounter (ie. the one I have encountered so far) is how to get the results of the intervening calculation (the new independent variable) which occur in the ML model back to the dataset to use in the model statement.
Any insight that users can provide is appreciated.
Regards,
Jonathan Alevy
*
* 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/
