Causal Inference & Experimentation

Designing credible evidence for business, product, policy, and operational decisions.

Focus

I use causal inference to help answer intervention questions: what will happen if an organization changes a price, launches a feature, sends a retention offer, targets a policy, reallocates spend, or changes an operational process?

Strengths

• Counterfactual framing and estimand definition.
• DAGs, assumptions, and adjustment strategies.
• A/B testing and experiment readouts.
• Difference-in-differences, regression discontinuity, instrumental variables, and synthetic control.
• Matching, weighting, doubly robust estimation, and heterogeneous treatment effects.
• Sensitivity analysis and decision-focused communication.

Applied Domains

• Pricing and promotion incrementality.
• Customer retention and targeting policy.
• Product experimentation.
• Marketing measurement.
• Health, education, and public policy analytics.
• Operations and resource allocation.

Evidence On This Site

• Causal inference projects
• Lecture notes
• Python causal package tutorials