ml-causal

Automate ML-based causal inference to estimate treatment effects in econometrics.

6|1|Updated Apr 3, 2026

Search Tags:#dml #machine-learning #causal-inference #lasso #causal-forest #cate #grf

Ready to run

npx skills add https://github.com/zhouziyue233/great-econometrics --skill ml-causal-zhouziyue233

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: ml-causal
Source: https://github.com/zhouziyue233/great-econometrics/tree/main/skills/ml-causal
Command: npx skills add https://github.com/zhouziyue233/great-econometrics --skill ml-causal-zhouziyue233

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

ML-based causal inference methods empower econometric analysis when there are many controls and suspected heterogeneity. This skill helps users estimate causal effects reliably by combining machine learning for nuisance components with rigorous identification strategies.

Core Features & Use Cases

Double/Debiased Machine Learning (DML) for partially linear models to estimate treatment effects with high-dimensional controls.
Causal Forests (GRF) for heterogeneous treatment effects and CATE estimation.
LASSO-based variable selection (Post-LASSO) to identify relevant controls before estimation.
BLP/CLAN analyses for assessing how CATE varies with observables and identifying subgroups.
AIPW / DR Learners for doubly robust estimation and model-agnostic robustness.

Quick Start

Provide a dataset with outcome, treatment, and covariates, then ask the AI to run a DoubleML analysis to estimate the ATE.

Frequently Asked Questions about ml-causal

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I estimate treatment effects with high-dimensional controls using Double Machine Learning?▼

Double Machine Learning estimates treatment effects by using ML to handle high-dimensional nuisance parameters. You provide outcome, treatment, and covariate data, then run a DML analysis to obtain valid causal inference through cross-fitting.

Can I use causal forests to estimate heterogeneous treatment effects?▼

Causal forests are used to estimate heterogeneous treatment effects and calculate the CATE. This skill applies generalized random forests (GRF) to identify how treatment effects vary across different subgroups within your dataset.

What is the best way to select relevant control variables before causal inference?▼

LASSO-based variable selection identifies relevant controls before causal estimation. You can apply Post-LASSO techniques to reduce high-dimensional datasets to a subset of meaningful variables, ensuring robust econometric analysis.

Does cross-fitting ensure valid inference for ML-based causal inference?▼

Cross-fitting ensures valid inference by splitting data to separate the ML nuisance estimation from the causal parameter estimation. This technique prevents overfitting and delivers honest estimates for treatment effects.

How do BLP and CLAN analyses assess treatment effect heterogeneity?▼

BLP and CLAN analyses assess heterogeneity by evaluating how the CATE varies with observable characteristics. BLP tests predictiveness, while CLAN identifies specific subgroups with distinct causal effects.

When should I use AIPW for doubly robust estimation in econometrics?▼

AIPW provides doubly robust estimation when you want model-agnostic robustness against misspecification. It combines outcome and treatment models to ensure consistent causal parameter estimation even if one model is incorrect.