Generalized Methods of Moments for Data Integration Across Disparate Studies
Nilanjan Chatterjee, PhD Bloomberg Distinguished ProfessorDepartment of Biostatistics, Bloomberg School of Public HealthDepartment of Oncology, School of MedicineJohns Hopkins University
In this talk, I will first review generalized method of moments as a unified framework for data integration across disparate studies. I will then discuss some more recent advances we have made in implementing this methodology for various types of real data applications,
including combining individual-level data from a main study and summary-statistics information from one or more external studies, improving computational scalability for the analysis of large-scale genome-wide association studies and development of high-dimensional predictive risk models.>/br>
This is joint work with Prosenjit Kundu, PhD and Ruzhang Zhao, PhD Candidate