EUVIMEDScopri EUVIMED
Open access · CC BY 4.0

Optimal Two‐Phase Sampling Designs for Generalized Raking Estimators With Multiple Parameters of Interest

Jasper B Yang iD, Bryan E Shepherd iD, Thomas Lumley iD, Pamela A Shaw iD

Statistics in Medicine · 2026

DOI: 10.1002/sim.70786 ↗

At a glance

Journal article in Statistics in Medicine, 2026. Written by Jasper B Yang and 3 co-authors. Institution: Department of Biostatistics, University of Washington, Seattle, Washington, USA. Major MEDLINE topics: Electronic Health Records, Models, Statistical. 47 references. The full text is freely available (open access, CC BY 4.0). Funded by: NIAID NIH HHS, U.S. National Science Foundation Graduate Research Fellowship Program, NIH HHS.

AbstractSubjectsDati bibliograficiCiteRelated studies

Abstract (original language)

Large observational datasets, including those derived from electronic health records, are a valuable resource for medical research but are often affected by missingness, measurement error, and misclassification. Two‐phase sampling with generalized raking (GR) estimation is an efficient and robust approach to statistical inference in such settings. In this approach, variables that are unavailable or measured with error in a large Phase 1 cohort are obtained with higher‐quality measurements in a Phase 2 subsample. Previous research has studied optimal Phase 2 sampling designs for inverse‐probability‐weighted (IPW) estimators in nonadaptive, multiparameter settings, and for GR estimators in single‐parameter settings. In this work, we extend these results by deriving optimal adaptive, multiwave sampling designs for IPW and GR estimators when multiple parameters are of interest. We propose several practical allocation strategies, examine the design choices that arise when implementing them, and evaluate their performance through extensive simulations and an example using data from people living with HIV attending the Vanderbilt Comprehensive Care Clinic (VCCC). Our results show that independently optimizing allocation for each parameter improves efficiency over traditional case‐control sampling. We also derive an integer‐valued, A‐optimal allocation method that typically outperforms independent optimization. Notably, we find that optimal designs for GR can differ substantially from those for IPW, and that this distinction can meaningfully affect estimator efficiency in the multiple‐parameter setting. These findings offer practical guidance for future two‐phase studies involving incomplete or error‐prone data.

Subjects

MeSH terms (MEDLINE)

Electronic Health RecordsModels, StatisticalHumansComputer SimulationHIV InfectionsObservational Studies as TopicProbabilityResearch DesignSampling Studies

Author keywords

A‐optimalityNeyman allocationgeneralized rakingmeasurement errortwo‐phase sampling

Authors and institutions

  • Department of Biostatistics, University of Washington, Seattle, Washington, USA.
  • Department of Biostatistics, Vanderbilt University, Nashville, Tennessee, USA.
  • Department of Statistics, University of Auckland, Auckland, New Zealand.
  • Biostatistics Division, Kaiser Permanente Washington Health Research Institute, Seattle, Washington, USA.

ORCID profiles: Jasper B Yang, Bryan E Shepherd, Thomas Lumley, Pamela A Shaw

Funding

  • NIAID NIH HHS
  • U.S. National Science Foundation Graduate Research Fellowship Program
  • NIH HHS
  • National Institutes of Health

Cite

APA

Yang, J. B., Shepherd, B. E., Lumley, T., & Shaw, P. A. (2026). Optimal Two‐Phase Sampling Designs for Generalized Raking Estimators With Multiple Parameters of Interest. Statistics in Medicine. https://doi.org/10.1002/sim.70786

Vancouver

Yang JB, Shepherd BE, Lumley T, Shaw PA. Optimal Two‐Phase Sampling Designs for Generalized Raking Estimators With Multiple Parameters of Interest. Statistics in Medicine. 2026. doi:10.1002/sim.70786

BibTeX
@article{Yang2026,
  author = {Yang, Jasper B and Shepherd, Bryan E and Lumley, Thomas and Shaw, Pamela A},
  title = {Optimal Two‐Phase Sampling Designs for Generalized Raking Estimators With Multiple Parameters of Interest},
  journal = {Statistics in Medicine},
  year = {2026},
  doi = {10.1002/sim.70786}
}
Start for free →