05. AI for Causal Inference
AI
LLMs
Causal Inference
Lecture Notes
How LLMs, RAG, agents, evaluation, and automation can support causal analysts without replacing identification discipline.
This course treats AI as an assistant to the causal analyst: translating business questions, drafting estimand cards, critiquing DAGs, retrieving domain knowledge, generating code, creating reports, and stress-testing AI outputs.
Notebook links open rendered HTML pages generated from the source notebooks under notebooks/lectures/. Code is visible by default; rendering is configured not to execute live notebook code, so local LLM or GPU-heavy cells are not triggered during website builds.
Notebook Sequence
- 00. Getting a Local LLM Running
- 01. AI-Assisted Causal Workflow
- 02. LLM Basics for Causal Analysts
- 03. Turning Business Questions into Causal Questions
- 04. Estimand Cards and Causal Design Documents
- 05. AI-Assisted DAG Brainstorming
- 06. DAG Critique, Variable Roles, and Backdoor Paths
- 07. RAG for Causal Domain Knowledge
- 08. Literature Synthesis for Causal Assumptions
- 09. Dataset Profiling with AI
- 10. Detecting Bad Controls, Post-Treatment Variables, and Leakage
- 11. Synthetic Data Generation for Causal Teaching
- 12. Simulation Labs for Assumption Stress Testing
- 13. AI-Assisted Method Selection
- 14. AI-Assisted Causal Code Generation
- 15. Automating Balance, Overlap, and Diagnostic Reports
- 16. AI for Sensitivity Analysis
- 17. AI for Experiment Design and Power Planning
- 18. AI for Quasi-Experiment Design
- 19. Causal Report Generation with LLMs
- 20. Causal Analysis Agent
- 21. Multi-Agent Causal Review Workflow
- 22. Evaluating AI Outputs in Causal Workflows
- 23. Hallucination and Failure Modes in AI Causal Analysis
- 24. Capstone AI-Assisted Causal Project
How To Read This Track
- Work through the notebooks in order if you want the full course arc.
- Treat each notebook as a lecture plus lab: read the discussion, inspect the code, and rerun locally when you want to experiment.
- For AI-heavy notebooks, expect some brittleness when live model calls are enabled; that instability is part of the course material rather than something hidden from the reader.
The .ipynb sources remain in the matching folder under notebooks/lectures/.