Research map / Probabilistic and scientific modeling / Treatment effects
Estimation and Optimization of Composite Outcomes
Daniel J. Luckett, Eric B. Laber, Siyeon Kim, Michael R. Kosorok
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| increase sample size from 100 to 500 | pseudo-likelihood omega estimation · Fixed utility, rho 0.60, omega 0.25 | error rate: 0.10 (0.11) -> 0.04 (0.04) | Improved |
| raise probability of optimal treatment rho from 0.60 to 0.80 | pseudo-likelihood estimation · Fixed utility, n=100, omega 0.25 | error rate: 0.10 (0.11) -> 0.03 (0.03) | Improved |
| use estimated-utility policy instead of standard of care | proposed method · Fixed utility sims, n=100, omega 0.25, rho 0.60 | policy value: 0.39 (0.23) -> 1.75 (0.29) | Improved |
| use estimated utility policy instead of single-outcome Y only policy | proposed method · Fixed utility sims, n=100, omega 0.75, rho 0.80 | policy value: 1.76 (0.08) -> 1.89 (0.07) | Improved |
| let rho vary with covariates (ablation) | proposed method · Variable rho sims | RMSE of beta: 1.32 (0.49) -> 0.43 (0.14) | Improved |
| allow patient-specific utility, increase n from 100 to 500 | proposed method · Variable utility sims | RMSE of theta: 2.42 (0.99) -> 1.68 (0.74) | Improved |
| increase n from 100 to 500 in variable utility sims | proposed method · Variable utility sims | error rate: 0.15 (0.08) -> 0.08 (0.04) | Improved |
| bootstrap test for utility heterogeneity, n=500 | parametric bootstrap · Homogeneity test | power against H2: 0.272 -> 0.682 | Improved |
| bootstrap test type I error with n=500 | parametric bootstrap · Homogeneity test | type I error: 0.066 at n=500, nominal 0.05 | No clear effect |
| let probability of optimal treatment vary between patients (ablation) | proposed method · STEP-BD SUM-D and SUM-M | value improvement: 0.1% -> 3.1% | Improved |
| let utility function vary between patients (ablation) | proposed method · STEP-BD SUM-D and SUM-M | value improvement: 3.1% -> 7.1% | Improved |
| let utility function vary between patients (ablation) | proposed method · STEP-BD SUM-D and side effect | value improvement: 6.9% -> 9.3% | Improved |
Closest papers by meaning
- An effective framework for estimating individualized treatment rules (NeurIPS 2024 · similarity 0.87)
- Doubly Robust Fusion of Many Treatments for Policy Learning (ICML 2025 · similarity 0.85)
- Assessing the robustness of heterogeneous treatment effects in survival analysis under informative censoring (NeurIPS 2026 · similarity 0.85)
- Doubly robust identification of treatment effects from multiple environments (ICLR 2025 · similarity 0.84)
- Defining Expertise: Applications to Treatment Effect Estimation (ICLR 2024 · similarity 0.84)
- PairNet: Training with Observed Pairs to Estimate Individual Treatment Effect (ICML 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
