Posts by Collection

scikit-tree

Cython-based library for rapid experimentation with tree methods

Implementations

Minimal didactic implementations of common algorithms in statistics and applied math

Semi-supervised Propensity Score Methods

Simulation studies on the use of unlabeled data in causal inference

The Relationship between SHAP and Functional ANOVA

Implementations and simulation studies that detail the links between SHAP and the functional ANOVA, including possible modifications to SHAP

Semi-supervised learning and the question of true versus estimated propensity scores

arXiv, 2020

Whether (and when) to use unlabeled data in causal inference via the propensity score.

Recommended citation: Herren, A., & Hahn, P. R. (2020). Semi-supervised learning and the question of true versus estimated propensity scores. arXiv preprint arXiv:2009.06183. https://arxiv.org/abs/2009.06183

Statistical Aspects of SHAP: Functional ANOVA for Model Interpretation

arXiv, 2022

Connections between SHAP and the design of experiments and sensitivity analysis literature.

Recommended citation: Herren, A., & Hahn, P. R. (2022). Statistical Aspects of SHAP: Functional ANOVA for Model Interpretation. arXiv preprint arXiv:2208.09970. https://arxiv.org/abs/2208.09970

Feature selection in stratification estimators of causal effects: lessons from potential outcomes, causal diagrams, and structural equations

arXiv, 2022

Insights on Feature Selection for Treatment Effect Estimation.

Recommended citation: Hahn, P. R. & Herren, A. (2022). Feature selection in stratification estimators of causal effects: lessons from potential outcomes, causal diagrams, and structural equations. arXiv preprint arXiv:2209.11400. https://arxiv.org/abs/2209.11400

Feature selection for learning causal effects: a finite-sample, stratification estimator perspective

Published: August 12, 2022

I presented my ongoing work on feature selection in causal inference at the 2022 NBER-NSF Seminar on Bayesian Inference in Econometrics and Statistics (SBIES). Slides from my talk are available here