.-
help for ^mclgen^                                                 (STB-56: sbe37)
.-

Stata programs for multinomial conditional logit models
-------------------------------------------------------

    ^mclgen^ depvar

The argument depvar is required. A categorical response factor with a
maximum of 12 levels must be specified for use in the MCL model.


Remarks
-------

Note that the ^mclgen^ program will ^modify the data^, and that the data
should therefore be ^saved^ before running it. An MCL model is estimated
on a "person/choice" file. In the "person/choice" file, each respondent
has a separate record for each category of the response variable. A
stratifying variable indexes respondents and a dichotomous variable
indicates which record corresponds with the repondent's actual choice. 

For a response factor with 5 levels, the dataset is therefore expanded
to 5 times the number of cases. The response factor specified in mclgen
indexes the response options for each respondent. ^mclgen^ also creates
^__strata^ and ^__didep^, the stratifying variable and dichotomous
variable for use by @mclest@.

When the model is estimated by @clogit@, the dichotomous variable
becomes the dependent variable and the stratifying variable is specified
in the ^strata()^ option. The main effects of the response factor
correspond with the intercept of a multinomial logistic model.
Interactions of the response factor with independent variables
correspond with the effects of these independent variables. These
effects can be created using @desmat@ (STB-52).

If the response factor has a fixed reference category, the log
likelihood, estimates and standard errors will be exactly the same as a
model estimated with @mlogit@. However, this procedure followed here
allows much more flexibility in imposing restrictions on the response
factor.

Author
------

    John Hendrickx
    Nijmegen Business School, University of Nijmegen
    P.O. Box 9108
    6500 HK Nijmegen
    The Netherlands
    J.Hendrickx@@mailbox.kun.nl