Hey everyone, I have a bit of trouble with plotting my continuous interactions using the marginsplot command after xt logit. The coefficient for the interaction comes up positive, but when I use margins and then marginsplot, the actual graph indicates a negative effect, whcih is quite confusing. Variables are coded correctly, I have checked that. I presume it must be a flaw in my margins command. I use:
margins, predict(pu0) at(var1=(x (x) x) var2=(x (x) x))
I have also tried:
margins c.var1#c.var2
But Stata gives me an error message 'variable c1 not found r(111)'
Any suggestions are very much appreciated. Thanks very much in advance for your help with this.