iptw

Nonparametric estimators of the average treatment effect with doubly-robust confidence intervals and hypothesis tests

Horvitz-Thompson estimator for RCTs, with Joel Middleton

Comparison of vedolizumab and tofacitinib via inverse propensity of treatment weighting

An end-to-end causal-inference workflow using synthetic observational longitudinal data: propensity-score weighting, balance diagnostics, GEE, and weighted survival analysis.

Analysis of Parkinson’s patients to explore how dopaminergic medication dose (LEDD) affects gait and fall risk. Combining causal inference (IPTW) with machine learning (XGBoost, logistic models), we reveal that high LEDD improves some gait metrics but increases fall risk, highlighting a clinical trade-off.

Code for Harnois-Leblanc, Rifas-Shiman, Switkowski et al. (2025) Estimating sex-specific population-level effects AJE

Synthetic NSCLC real-world-data pipeline with known ground truth: line-of-therapy derivation, IPTW, immortal-time-bias demo, 10-seed Monte-Carlo truth recovery (Python)

Reproducible clinical/RWE causal-inference workflow estimating the effect of early right heart catheterization on 30-day mortality using IPTW, doubly robust AIPW, overlap diagnostics, sensitivity analysis, and Python.

Config-driven Quarto/Rmd workspace for survey-weighted observational papers: survey design, IPTW, LCA, network analysis and reporting skeleton. Clone once per paper.

New-user active-comparator study on Medicare claims: CABG vs PCI with propensity score weighting, matching, balance diagnostics and competing-risks survival analysis. SQL cohort construction over CMS DE-SynPUF.

Official R implementation of the preoperative dual-score framework distinguishing recurrence risk from death without recurrence to assess resection benefit in hepatocellular carcinoma.