ASA Connect

Dear Colleagues, 

This is a reminder that ASA Statistical Learning and Data Science Section will have its next webinar this Thursday, May 8.  Prof. Peter Song from University of Michigan will present a novel functional data modeling framework developed to evaluate the influence of physical activity on biological age or age acceleration.  Hope to see you there!  

Title:                        Functional accelerometer data analysis via mixed integer optimization

Speakers:               Dr. Peter Song, Department of Biostatistics, School of Public Health, University of Michigan

Date and Time:      May 8, 2025, 3:00 to 4:30 pm Eastern Time

Registration Link:  ASA SLDS Webinar Registration Link [eventbrite.com]

Abstract:                Accelerometer sensors capture high-frequency time series data that reflect rich functional features of physical activity. In this talk, we present a novel functional data modeling framework developed to evaluate the influence of physical activity on biological age or age acceleration. Our approach leverages a homogeneity pursuit approach based on mixed integer optimization (MIO) to identify critical time windows of activity that significantly impact biological aging. The MIO framework enables simultaneous operations of change point detection and parameter estimation, enhancing overall error control and estimation precision. The resulting functional parameter estimate enjoys desirable interpretability. We demonstrate the effectiveness of the proposed methodology through simulation studies and real-world data applications on the association between physical activity and skin age among adolescents. This is a joint work with Margaret Banker and Leyao Zhang.

Presenter:              Dr. Song is Professor of Biostatistics at the University of Michigan School of Public Health, Ann Arbor. He received his PhD in Statistics from the University of British Columbia, Vancouver, Canada in 1996. He has published over 230 peer-reviewed papers and graduated 26 PhD students and trained 6 postdoc research fellows. Dr. Song's current research interests include data integration, distributed inference, high-dimensional data analysis, longitudinal data analysis, mediation analysis, and spatiotemporal modeling with applications in nephrology, chronic disease epidemiology, environmental health sciences, and nutritional sciences. He is IMS Fellow, ASA Fellow and Elected Member of the International Statistical Institute. Dr. Song now serves as Area Editor of the Annals of Applied Statistics (Medicine, EHR and Smart Health), Associate Editor of the Journal of American Statistical Association and the Journal of Multivariate Analysis, and Statistics Reviewer for Kidney International.