> For at each one of the dummy groups in your example oaxaca cannot find
> such a reference dummy variable.
"For at least one ..." I wanted to say.
ben
On Mon, Nov 24, 2008 at 9:36 PM, Ben Jann <[email protected]> wrote:
> A dummy variable for the reference category must exist in the data. So
> if, say, condnum has 9 categories, 1 through 9, you need nine dummies,
> one for each category. In the main model, you can only specify eight
> dummies (to prevent perfect collinearity), but if you use the
> categorical() option then the 9th dummy has to exists and you have to
> identify it.
>
> For at each one of the dummy groups in your example oaxaca cannot find
> such a reference dummy variable.
>
> ben
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
