Reproduce “A Guide to Estimating Conditional Average Treatment Effects in Competing Risks Settings” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Daniel Klippert, Sarah Friedrich, Markus Pauly
This research systematically evaluates six meta-learning algorithms for estimating Conditional Average Treatment Effects (CATEs) in time-to-event settings with competing risks, providing practical guidance for method selection and introducing an open-source R package for implementation.
Key insights
- The complexity of the true CATE function and the degree of treatment imbalance are primary drivers of meta-learner performance.
- Elastic net regression models for CATE functions generally exhibit robust performance and often outperform random forest-based approaches.
- S- and T-Learners show increased sensitivity to nuisance model misspecification, while R- and DR-Learners with elastic net CATE models demonstrate greater robustness.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.