Quantcast
Channel: Statalist
Viewing all articles
Browse latest Browse all 65123

Graphing predicted probabilities for continuous variables on distinct categories (interactions)

$
0
0
Dear Statalist members,

I am working on a project where the main analysis focuses on logit/probit with interaction terms between one binary covariate (x1) and other covariates that are continuous (x2, x3, x4).

I would like to add graphs which would enrich a lot the analysis and help me with the interpretation.

I would like to do something like the graphs in Greene (2010, Testing hypotheses about interaction terms in nonlinear models, Figures 1 and 4).
Array

Array


I found this document www.stata-press.com/books/mlmus3_ch10.pdf, which is not exactly like Greene, but which helps me making my point.


Array

The code they suggest is this one:

twoway (line predicet_probability x2 if x1=0, sort)(line predicet_probability x2 if x1=1, sort lpatt(dash)).

But if I do this, what I get is:

Array


Can someone offer me some guidance on how to do it properly?



Best,

Mari



Viewing all articles
Browse latest Browse all 65123

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>