A central problem of interest to social scientists, businesses, and government agencies alike is the evaluation of the effects of a treatment or policy intervention. Causal Inference is the statistical field of study concerning this problem, and it boasts a mature literature that contains numerous sophisticated methodology and recommendations.
Unfortunately, much of this literature has so far proved difficult to access for non-specialists due to its technical nature and the high implementation costs associated with the more advanced methods. The software package Causal Inference in Python, or Causalinference in short, is an attempt to bridge this gap by exposing the statistical tools in an easy-to-use interface.
This series of blog posts will illustrate the use of Causalinference, as well as provide high-level summaries of the underlying econometric theory with the non-specialist audience in mind. For those who are already familiar with the literature, I recommend referring to the much terser vignette paper. These blog posts can be thought of as an expanded but less formalized version of that material.