上海财经大学 > 科学研究 > 学术交流 > 学术报告
  • 统计与管理学院2017年学术报告第46

     

    【主  题】Using Ranked Set Sampling with Generalized Linear Mixed Models for Clustered Data

    【报告人】Xinlei Wang,教授

    Southern Methodist University

    【时  间】 2017年06月26日(星期一)10:00-11:00

    【地  点】 上海财经大学统计与管理学院大楼1208室

    【摘  要】We incorporate ranked set sampling (RSS), a cost-efficient method, into cluster-randomized designs (CRDs), for potential improvement in estimation and detection of treatment or intervention effects. Outcome data in cluster-randomized studies typically have nested structures, where Generalized Linear Mixed Models (GLMMs) become a natural choice for data analysis. However, nearly all theoretical/methodological developments in RSS in the past were within the structure of one-level models. Thus, implementation of RSS at one or more levels of a GLMM will require development of new theory and methods. In this study, we propose new RSS-structured CRDs with improved estimators and testing procedures for both continuous and binary responses.

    Related Publications:

    Wang, X., Lim, J., and Stokes, S. L. (2016), “Using Ranked Set Sampling with Cluster Randomized Designs for Improved Inference on Treatment Effects”. Journal of the American Statistical Association111(516), 1576-1590.

    Wang, X., Ahn, S., and Lim, J. (2017), “Unbalanced Ranked Set Sampling in Cluster Randomized Studies”. Journal of Statistical Planning and Inference187, 1-16.

    Ahn, S., Wang, X., and Lim, J. (2017), “On Unbalanced Group Sizes in Cluster Randomized Designs Using Ranked Set Sampling”. Statistics and Probability Letters123, 210-217.

    Wang, M. (2017), “Using Ranked Set Sampling with Binary Outcomes in Cluster Randomized Designs”. Ph.D. Dissertation, Department of Statistical Science, Southern Methodist University.

    【嘉宾简介】 Dr. Wang received her B.S. degree in Automatic Control from University of Science and Technology of China in 1997, and received her PhD degree in Decision Science and Statistics from University of Texas at Austin in 2002. Currently, she is a Professor of Statistics at Southern Methodist University, USA. She conducts methodological research in multiple areas of Statistics, including Bayesian methods and applications, integrative analysis, meta-analysis, and order-related design, theory and inference.

    Dr. Wang has published 40+ methodological papers in Statistics including those in top statistical journals such as Journal of the American Statistical Association (JASA), Biometrika, Biometrics, Bioinformatics, etc. In the past, she has been the dissertation advisor of 12 Ph.D. students and 2 Master's students, as well as serving as a committee member for 13 Ph.D. graduates. Her research has been supported by multiple NSF and NIH grants, where she is the PI or Co-PI.

    【邀请人】 李涛