Time: Monday, November 13, 2:00 PM-3:00 PM CDT
Location: 4th Floor Conference Room 400 in the Doctors Office Building at 66 N. Pauline Street, Memphis, TN 38105.
Please park in the multi-story parking garage adjacent to the Doctors Office Building, and bring your parking ticket with you so we can validate it.
ZOOM Virtual Room Connection: Register in advance for this meeting
Seminar Website: https://www.eventcreate.com/e/biostatisticsseminar
Speaker Bio: https://www.memphis.edu/publichealth/contact/faculty_profiles/mou.php
Generalized kernel machine regression
Xichen Mou, Ph.D.
School of Public Health, University of Memphis
Division of Epidemiology, Biostatistics, and Environmental Health
Kernel Machine Regression (KMR) serves as a nonparametric regression approach fundamental in numerous scientific domains. By utilizing a map determined by the kernel function, KMR transforms original predictors into a higher-dimensional feature space, simplifying the recognition of patterns between outcomes and independent variables. KMR is invaluable in studies within the biomedical and environmental health sectors, where it aids in identifying crucial exposure points and gauging their impact on results. In our study, we introduce the Generalized Bayesian Kernel Machine Regression (GBKMR) which integrates the KMR model within the Bayesian context. GBKMR not only complements the conventional KMR but also suits a range of outcome data, from continuous to binary and count data. Simulation studies confirm GBKMR’s superior precision and robustness. We further employ this method on a real data set to pinpoint specific cytosine phosphate guanine (CpG) locations correlated with health-related outcomes or exposures.