[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: interaction term in probit (inteff problem)
At 10:24 AM 6/23/2006, Kai-Wen Cheng wrote:
It doesn't sound to me like you can do it. The help file says it is
for "when two variables have been interacted." In your setup, you've
got 9 vars (year2 through year10) interacted with college. I guess
you could include those terms beginning with variable 5, but they'll
be treated as continuous vars and the program won't know they are
By the way, I have another question about inteff commands...
I am using 10 years data and I have to figure out the interaction effects
between college(a dummy variable) and every year. That is, I need to
figure out the interaction effects of year2*college, year3*college,
In Norton, Wang, Ai's paper, it said put dummy1 in the second position and
dummy 2 in the third and then dummy1*dummy2 in the fourth...But how about
getting more than one interaction terms? how to apply the commands to my
If it seemed legit to treat year as a continuous var, then you would
just have year and college and you could use the program.
I suspect this would not be legitimate, but I suppose you could keep
on rerunning the program rotating which vars are in positions 2, 3 and 4.
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
WWW (personal): http://www.nd.edu/~rwilliam
WWW (department): http://www.nd.edu/~soc
* For searches and help try: