Stan and BART for Causal Inference: Estimating Heterogeneous Treatment Effects Using the Power of Stan and the Flexibility of Machine Learning

Mathematics

• Regression Tree 100%
• Causal Inference 92%
• Machine Learning 74%
• Treatment Effects 73%
• Flexibility 64%
• Nonlinear Response 64%
• Response Surface 58%
• Average Treatment Effect 34%
• Performance 26%
• Standard Model 25%
• Computational Efficiency 24%
• Parametric Model 23%
• Model 22%
• Subgroup 15%
• Observation 14%
• Range of data 13%
• Demonstrate 13%
• Estimate 10%
• Form 9%

Physics & Astronomy

• machine learning 86%
• inference 75%
• flexibility 61%
• estimating 60%
• subgroups 20%
• performance 14%
• estimates 9%

Engineering & Materials Science

• Machine learning 49%
• Computational efficiency 16%