Jiayin Zheng, PhD
Dr. Zheng is an Assistant Professor of Biostatistics in the Department of Biostatistics, Epidemiology and Informatics (DBEI) at the Children’s Hospital of Philadelphia and the Perelman School of Medicine at the University of Pennsylvania. Dr. Zheng is the associate director for Biostatistics at the Clinical Futures, a Research Institute Center of Emphasis at Children’s Hospital of Philadelphia.
Dr. Zheng earned a Ph.D. in Probability and Statistics from Peking University in Beijing, China under the supervision of Shuyuan He and a B.S. in Statistics from Zhejiang University in Hangzhou, China. Before joining the University of Pennsylvania and the Children's Hospital of Philadelphia, he served as a Staff Scientist in the Biostatistics Program at Fred Hutchinson Cancer Center. He also obtained postdoctoral training at Duke University and Fred Hutchinson Cancer Center.
His research focuses on the development of novel statistical methods to address important scientific problems and the application of sound statistical approaches to experimental/real-world data to facilitate knowledge discovery and improve decision-making in public health and clinical medicine. He enjoys working on extensive collaborative projects in various research areas including, but not limited to, colorectal cancer, gastroenterology, pediatric research, osteoporosis, cardiology, and infectious diseases. In his free time, he enjoys hiking and playing soccer.
Content Area Specialties
Colorectal cancer, gastroenterology, pediatric research, osteoporosis, cardiology, and infectious diseases.
Methods Specialties
Motivated by his extensive collaborative experiences, his methodological research spans several innovative research areas, including data integration, design and analysis of multiple-phase sampling (e.g., case-control studies), observational data analysis, risk prediction modeling and evaluation, and survival analysis.
Recently, his methodological research interests have focused on data integration, with an emphasis on leveraging external summary information with internal individual data to gain efficiency in parameter estimation and reduce bias for model generalizability/transportability. The motivation stems from his postdoc training at Duke University, where he noticed the limitations of regional/local/hospital-level data, e.g., inadequate sample size leading to insufficient power/efficiency, and biased sampling schemes making results under-represent the target population. He is also interested in developing predictive models using individual participant data meta-analysis and is leading a project to develop a web-based colorectal cancer risk prediction tool based on common genetic risk factors and environmental and lifestyle risk factors.