[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Interaction with squared variable
Giuseppe Cappelletti <[email protected]> has discovered a bug
when specifying interactions with squared terms in -oprobit-. He writes:
> I would like to estimate an orderd probit model with the following
> specification:
> Prob(y=1) = x + a*gender + b*x + c*x^2 + d*gender*x^2
> I tried the following command:
> oprobit y ///
> i.gender  ///
> c.x  ///
> c.x#c.x ///
> i.gender#(c.x#c.x)
> but it does not work. After the estimation I would like to compute the
> marginal effect (considering both the squared and interaction term).
This bug is triggered by the inclusion of both a squared term and its
interaction with a factor variable. We will fix this in an update.
--Bobby
[email protected]
*
* 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/