Topic: Time-Varying and High-Dimensional Confounding

Rare Features and Separation in Propensity Models

Prioritizing Proxies With the HDPS Algorithm

Prognostic Covariates and Instrumental Features

Overlap and Weight Instability in High-Dimensional Scores

Nonlinear Terms and Interactions in High-Dimensional Scores

Finding Residual Imbalance Within Feature Blocks

Encoding Diagnosis and Procedure Histories

Effective Sample Size Under High-Dimensional Weights

Dimension Reduction With Causal Structure Constraints

Deriving the HDPS Bias-Reduction Heuristic