causal-inference-analysis

To implement a difference-in-differences model in Python, you can use the `linearmodels` or `pyfixest` libraries. This Skill provides code templates for setting up two-way fixed effects models, running event study analyses, and generating publication-ready regression tables for staggered adoption designs.

To test instrumental variables assumptions in Python, use libraries like `linearmodels` or `doubleml`. This Skill guides you through essential diagnostic checks, including calculating first-stage F-statistics to assess instrument strength and evaluating covariate balance for robust treatment effect estimation.

Yes, you can run regression discontinuity design diagnostics using the `rdrobust` and `rddensity` packages in Python. This Skill provides implementation examples for estimating treatment effects at cutoffs and testing identification assumptions specific to RDD methodologies.

To create synthetic control models for policy evaluation, use the `pysynth` or `synth_runner` dependencies. This Skill offers practical code templates for constructing weighted donor pools, estimating causal effects of new policies, and generating comparative event study plots.

Yes, this Skill supports Stata code generation for econometric analysis via the `pystata` dependency. It provides implementation guidance for Difference-in-Differences, Instrumental Variables, and Synthetic Control methods, alongside Python code and publication-ready output generation.

You should use `doubleml` or `econml` for causal inference when dealing with high-dimensional confounders or complex treatment effect heterogeneity. This Skill utilizes these dependencies to implement robust treatment effect estimation and assumption testing beyond standard linear regression approaches.