Hi y’all…
I’m a doctoral student gearing up for data science and economist roles that require strong causal inference skills. In our causal inference class, we primarily used STATA, but I’m curious about what software is most commonly used in the industry.
It depends on the use case. Most of the people I work with are more comfortable with python, so python is my default choice. The packages I use are primarily EconML and PyWhy, with a healthy dose of statsmodels and scipy. Occasionally we’ll work on projects that have a specific, tailormade R library, in which case I’ll use it. In some cases Bayesian modeling is more appropriate, in which case I’ll use STAN, but that was more common in my last job.
It really depends where you work and who you work with. In general I think python is a little more common in causal inference roles, but nowhere near as common as in roles that are more predictive in nature.
It depends on the use case. Most of the people I work with are more comfortable with python, so python is my default choice. The packages I use are primarily EconML and PyWhy, with a healthy dose of statsmodels and scipy. Occasionally we’ll work on projects that have a specific, tailormade R library, in which case I’ll use it. In some cases Bayesian modeling is more appropriate, in which case I’ll use STAN, but that was more common in my last job.
It really depends where you work and who you work with. In general I think python is a little more common in causal inference roles, but nowhere near as common as in roles that are more predictive in nature.