You're comparing apple pie and orange juice here (even worse than comparing apples and oranges). Pandas is an open-source library for people who know python programming. Stata is an expensive GUI-based suite for non-programmers. If you're not in university, government, or non-profit, Stata's single-CPU license is $840/year in the US. Even a single-CPU student license is $94/year.
The only people still using stata are 40+ year old economics professors and the poor lab students they force into obsolescence. Everybody else is using R (if you're into econometrics) or Python (if you're into ML).
Stata is still big in pharma. There are moves happening to R, but it is a conservative space where businesses do not want to rock the boat too much with the FDA. If nothing else as a Python user, I think the pinning/reproducibility story in R still has quite a ways to go.
