Hi Chuck, is it possible, using a specific frame, to make a twoway contour if my regress contains a var1^2, var1 and var2. You help would be very usefull for my work. Thanks. Gerar
It sounds like you want to create a contour plot, but it is unclear whether you want to plot the actual values of your variables or predicted values after fitting your model. Note that you can use *predict* to obtain predicted values, and then plot those resulting variables. If you would like, you can email us at tech-support@stata.com with more details about the graph you would like to create, and we would be happy to help.
Hi, whenever I put in my variable in it comes up as covariate not found, do you have any ideas how to go about this? Any help would be appreciated. Thanks
The variable you specify immediately after *margins* needs to be a factor variable or an interaction term. For example, in this video we fit a model with *regress bwt age i.smoke* . We were then able to type *margins smoke* . On the other hand, age is not a factor variable, so we would get an error if we were to type *margins age* . Please type *help margins* to learn more and feel free to email our Technical Support team at tech-support@stata.com for any other questions.
Hi I have a problem when I use "margins". Could you please have a look at the error message as the following? margins, at(meta1=(-3, 0)) default prediction is a function of possibly stochastic quantities other than e(b) what can I do to fix this problem? Thanks!
