Skip to content

Latest commit

 

History

History
19 lines (12 loc) · 1.17 KB

classical_methods.md

File metadata and controls

19 lines (12 loc) · 1.17 KB

The Classical Methods of Causal Inference

We have introduced and adapted some commonly used methods of causal inference from Econml, which can be well suited for task scheduling and execution in this platform.

Meta-Learners

Meta-Learners models the response target and estimates the treatment effect by quantifying the change in the target variable caused by the treatment. It contains three main methods: T-Learner, S-Learner and X-Learner.

The example: examples/openasce/inference/learner/metalearners.py

Doubly Robust Learning

DRLearner is a doubly robust estimation method based on two-stage estimation that can effectively estimate heterogeneity when there are confounders in the observed data, which can reduce bias effectively.

The example: examples/openasce/inference/learner/drlearner.py

Double Machine Learning

DML is a method used in studying Heterogeneous Treatment Effects (HTE) that allows for unbiased estimation of Average Treatment Effects (ATE) even when the estimation of the nuisance parameter W is biased, by estimating the moment residuals.

The example: examples/openasce/inference/learner/dml.py