Here is the code I used to make Figure 1. Don’t limit yourself to the two that I did here. Make some interaction plots, which can be a helpful visual. I’m going to share the code that I used and you should try it out with your models and your variables. The best way to learn this part is to just do it. What you see up there is (obviously) a combination of two graphs made into one, which is a thing you can do in Stata. The book is a companion to the Spost13 Stata commands (that you can use without the book, of course) but the text will help you know what you’re doing. And I highly recommend Long and Freese (2014) “ Regression Models for Categorical Dependent Variables Using Stata” as a go-to reference manual for Stata and nonlinear regression, with an emphasis on interpretation. And you’ll want to run it later to check for updates on all of it. What you’re doing is installing several excellent programs for post-estimation work in Stata, including several programs written by statistics pros Ben Jann and Scott Long. However you do it, make that a complete do-file that you have, run, and then save somewhere. On Windows, you can right-click and choose “save link as” or if you just click the link, it will open in a new tab, then you can copy it all into a blank do-file if you prefer. I recommend downloading this do-file and running it first. Edit: Instead of adding to this page, I redid this tutorial and made a much better one. This is just round one to get folks something to work with. This should get you started on visuals! I will add to this page so check back and refresh.
Graph combine stata how to#
This is not intended to be an exhaustive tutorial, but, rather, a sampling of how to make a few graphs for your (mostly) nonlinear regression models using some stata commands (margins and marginsplots) and some of Ben Jann’s programs (coefplot and grstyle in particular) while making use of Spost13.
This tutorial has been substantially improved upon and updated. This page has a short link: /coef1 will bring you back here.
0 kommentar(er)
