Research map / Learning theory and algorithms / Prediction and decision losses
Optimal testing using combined test statistics across independent studies
Lasse Vuursteen, Botond Szabo, Aad van der Vaart, Harry van Zanten
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 |
|---|---|---|---|
| combine local chi-square statistics across m trials instead of using a single trial | chi-square combined meta-analysis test · Synthetic many normal means, m=20, n=30, d=2..20 | ROC (TPR vs FPR) | Improved |
| use pooled data chi-square test instead of combining local chi-square statistics | chi-square combined meta-analysis test · Synthetic many normal means, m=20, n=30, d=2..20 | ROC (TPR vs FPR) | Improved |
| combine local chi-square statistics instead of directional statistics in high dimension | meta-analysis combination method · Synthetic, d=10 and d=20 | ROC (TPR vs FPR) | Improved |
| combine directional statistics instead of local chi-square statistics in low dimension | meta-analysis combination method · Synthetic, d=2 | ROC (TPR vs FPR) | Improved |
| add coordination between trials via shared randomness (coordinated directional vs uncoordinated directional) (ablation) | directional meta-analysis test · Synthetic, d=2 and d=5, small alpha | ROC (TPR vs FPR) | Improved |
| explore eight additional settings with larger d and m | all combination strategies · Synthetic appendix simulations | ROC (TPR vs FPR) | No clear effect |
Closest papers by meaning
- Integrative High Dimensional Multiple Testing with Heterogeneity under Data Sharing Constraints (JMLR 2021 · similarity 0.85)
- Asymptotically Optimal and Computationally Efficient Average Treatment Effect Estimation in A/B testing (ICML 2024 · similarity 0.83)
- Enhancing Statistical Validity and Power in Hybrid Controlled Trials: A Randomization Inference Approach with Conformal Selective Borrowing (ICML 2025 · similarity 0.82)
- Efficient Randomized Experiments Using Foundation Models (NeurIPS 2025 · similarity 0.82)
- Data Fusion for Partial Identification of Causal Effects (NeurIPS 2025 · similarity 0.81)
- Adaptive Data-Borrowing for Improving Treatment Effect Estimation using External Controls (NeurIPS 2025 · similarity 0.81)
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.
