Causal Inference & Experimentation
Moving beyond correlation to causation: Rigorous statistical methods that measure true business impact and guide strategic decisions.
Featured Case Study
Medicaid Expansion Impact Study
Rigorous causal analysis of the 2014 Medicaid expansion using DiD methodology. Analyzed healthcare access across 50 states, comparing treatment (expansion) and control (non-expansion) groups.
Key Finding:
2.1 percentage point reduction in cost-related healthcare delays
16% improvement in healthcare access for expansion states
fewer poor mental
health days/month
After correcting for selection bias
Exercise & Mental Health Study
IPW analysis revealing the true causal effect of exercise on mental health. Corrects for self-selection bias where healthier, higher-income individuals are more likely to exercise.
Selection bias impact:
Coming Soon
Promotion Threshold Analysis
Using RDD to measure the causal impact of promotional campaigns by exploiting arbitrary cutoffs in customer spending or engagement metrics.
Market Expansion Impact
Constructing synthetic controls to measure the causal impact of geographic expansion when randomized experiments aren't feasible.
Technical Expertise
Languages
Python, R, SQL, Julia
ML Frameworks
TensorFlow, PyTorch, Scikit-learn, XGBoost
MLOps
MLflow, Kubeflow, SageMaker, Vertex AI
Causal Methods
A/B Testing, DiD, RDD, IV, Synthetic Control
Let's Measure What Actually Matters
Whether you need to prove ROI, optimize experiments, or establish causation, I bring rigorous statistical methods to your toughest analytics challenges.
Connect With Me