In the 20th century, frequentist statistics dominated the field of statistics, but it didn’t fully address causal inference. Causal inference gained attention in the latter half of the century, thanks to advancements in statistical methods and the work of Judea Pearl. However, the data science community didn’t prioritize causal inference until around 2015. Causal inference helps businesses make better decisions by considering alternative outcomes. It uses methods like linear regression, causal graphical models, instrumental variables, difference in differences, synthetic control, and double ML. These methods allow researchers to estimate causal effects and understand the impact of interventions in various scenarios.
https://dzidas.com/ml/2023/10/15/blind-spot-ds/