[Limdep Nlogit List] nested logit

Bidisha Sayema lexshb at nottingham.ac.uk
Wed Dec 3 06:02:15 EST 2008

I am new in NLOGIT and having problem while estimating a model. I am trying to estimate a nested logit model where in the 1st stage respondents choose between 4 alternatives: wage employment, self employment, unemployment and other employment. In the next stage the wage employed choose among 3 alternatives: professional, skilled and unskilled. We dont consider any further choices for the other 3 categories. All my variables are choice-invariant (so all are Rh2)and are demographic and educational variables (many of them are dummies). I have 26 variables in the 2nd stage and 28 in the 1st stage. In both stages most of the variabl;es are common but only in 1st stage there are 2 additional variables which are not in 2nd stage. 
I am using the menu bar to estimate it and able to get estimates without specifying any IVs and with specifying the tree but without specifying the utility function in particular. So I am using the 26 variables only and all are interacted with ASCs. I was trying to use the standard Nlogit comman rather than the 2 step one.I want to know whether Nlogit is able to estimate such models ie. whether my results are making any sense. In certain cases it was not able to estimate my model which made me skeptical about my results.  In addition, how can I incorporate 2 different utility functions in these 2 stages and whether I should specify anythinmg about IVs. At the moment I am not specifying any. 
Any suggestion would be greatly appreciated. Many thanks in advance.
Best Regards

This message has been checked for viruses but the contents of an attachment
may still contain software viruses, which could damage your computer system:
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

More information about the Limdep mailing list