Skip to content
DraconiticResearch
Search Sources API
186,086 studies31 sourcesUpdated 11 min ago
PreprintarXiv licencearXiv

Prediction-Powered Data Fusion for Treatment Effect Estimation

  • Yonghan Jung
  • Shu Yang
Open at sourcePDFDOICite

Abstract

Randomized controlled trials (RCTs) identify treatment effects without confounding but are often small, whereas observational studies (OBS) are large but may be confounded. Many estimators combining a small RCT with a large OBS have been developed for the average treatment effect (ATE) and the conditional ATE (CATE). However, existing ATE estimators either make assumptions on the OBS or do not borrow enough power from them. The CATE has been studied less than the ATE. Existing CATE methods either assume the OBS are unconfounded, rely on a model of the confounding function, or accept bias in exchange for lower variance. We therefore propose a framework that, without special assumptions on the OBS, fuses the OBS and the RCT by preserving the unbiasedness of RCT-based estimation while borrowing power from the large OBS to boost precision. Applying this principle, we build an ATE estimator, AIPW-Fusion, with closed-form weights and confidence intervals, and two CATE learners, DR-Fusion and R-Fusion. Experiments corroborate our findings.

Related

  1. PPI++: Efficient prediction-powered inference
    Anastasios N. Angelopoulos, John C. Duchi, Tijana Zrnic
    The Annals of Applied StatisticsPeer-reviewedLicence not statedFull text26 citedStatisticsOpenAlex
  2. Beyond Point Predictions: Distribution-Informed Prediction-Powered Inference
    Maoyu Zhang, Jingfei Zhang, Xuming He
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  3. Covariance-Calibrated Prediction-Powered Inference
    Siyuan Chen, Yang Bai, Huichen Zhu
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  4. Which Metrics Save the Most Human Annotation? Prediction-Powered Evaluation and Meta-Evaluation
    Mingqi Gao, Anthony Sicilia, Weiyan Shi
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  5. ABC-Align: Prediction-Powered Alignment with Adaptive Bias Control
    Eric Frankel, Banghua Zhu, Sewoong Oh +1
    arXivPreprintarXiv licenceFull textStatisticsarXiv

Cite

BibTeXRIS
BibTeX
@misc{jung2026predictionpowered,
  title = {{Prediction-Powered Data Fusion for Treatment Effect Estimation}},
  author = {Yonghan Jung and Shu Yang},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.12332},
  eprint = {2610.12332},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.12332},
}
RIS
TY  - UNPB
TI  - Prediction-Powered Data Fusion for Treatment Effect Estimation
AU  - Yonghan Jung
AU  - Shu Yang
PY  - 2026
DA  - 2026/10/23
PB  - arXiv
DO  - 10.48550/arxiv.2610.12332
UR  - https://arxiv.org/abs/2610.12332
AB  - Randomized controlled trials (RCTs) identify treatment effects without confounding but are often small, whereas observational studies (OBS) are large but may be confounded. Many estimators combining a small RCT with a large OBS have been developed for the average treatment effect (ATE) and the conditional ATE (CATE). However, existing ATE estimators either make assumptions on the OBS or do not borrow enough power from them. The CATE has been studied less than the ATE. Existing CATE methods either assume the OBS are unconfounded, rely on a model of the confounding function, or accept bias in exchange for lower variance. We therefore propose a framework that, without special assumptions on the OBS, fuses the OBS and the RCT by preserving the unbiasedness of RCT-based estimation while borrowing power from the large OBS to boost precision. Applying this principle, we build an ATE estimator, AIPW-Fusion, with closed-form weights and confidence intervals, and two CATE learners, DR-Fusion and R-Fusion. Experiments corroborate our findings.
ER  -