st: Problem probing interaction

[Kris Anderson <andersok@reed.edu>]

Hi, all.

I am trying to plot a 2-way continuous vs continuous interaction in 
logistic regression using multiple imputation. I have successfully done 
this for 3-way interactions using different code. I found this code on 
the site but can't get it to run.

mim: logistic drinker gradeR sexR his_latR white C_ndmot C_socmot 
C_copemot C_enhmot C_confmot int_nsoc int_ncope int_nenh int_nconf
gen y1 = _b[C_socmot]*(-1.18) + _b[C_ndmot]*C_ndmot + 
_b[int_nsoc]*(-1.18)*C_ndmot + _b[_cons]
gen y2 = _b[C_socmot]*(0) + _b[C_ndmot]*C_ndmot + _b[ 
int_nsoc]*(0)*C_ndmot + _b[_cons]
gen y3 = _b[C_socmot]*(1.18) + _b[C_ndmot]*C_ndmot + _b[ 
int_nsoc]*(1.18)*C_ndmot + _b[_cons]
sort C_socmot
twoway (line y1 ) (line y2 C_socmot) (line y3 C_socmot), 
name(SocmotxNdm, replace)
gen p1 = exp(y1)/(1+exp(y1))
gen p2 = exp(y2)/(1+exp(y2))
gen p3 = exp(y3)/(1+exp(y3))
twoway (line p1 C_socmot) (line p2 C_socmot) (line p3 C_socmot), 
name(SocmotxNdm, replace)

I am getting an error message after the gen y1 line saying that "no 
variables defined." The code looks like my sample so am a bit confused.

Thanks,
Kris
*
*   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/