Unobserved Heterogeneity in oprobit/ologit Models

[Warren Gee]

Hello!

I have non-panel data and I’m trying to figure out if there is a way to analyze unobserved heterogeneity in either my oprobit or ologit models.

Thanks!

[Andrew Musau]

Unobserved heterogeneity is a pervasive problem in cross-sectional analyses. The only way to mitigate it is to add as many control variables as possible. You can argue that your control variables are sufficient, but there is no direct way to test this. In the future (assuming that your study is one of the first), other researchers may get access to panel data and will be able to control for it. If they reach the same conclusions as you, then your argument is validated. But in general, cross-sectional studies are considered to be descriptive and not causal.