Bayesian Filtering for Time-Varying Parameter Estimation in Biological Models
Andrea Arnold, Ph.D.
Mathematics
North Carolina State University
ABSTRACT
Many applications in the life sciences involve unknown system parameters that must be estimated using little to no prior information. In particular, some parameters may be known to vary with time with no known evolution model, yet may be subject to certain structural characteristics such as periodicity. We show how nonlinear Bayesian filtering techniques can be employed in this setting to estimate unknown, time-varying parameters, while naturally providing a measure of uncertainty in the estimation. Results are demonstrated using real-world data from several biological applications, including cardiovascular dynamics and the transmission of infectious diseases.
BIOGRAPHY
Dr. Andrea Arnold is a Postdoctoral Fellow with the Research Training Group in Mathematical Biology in the Department of Mathematics and the Center for Quantitative Sciences in Biomedicine at North Carolina State University (Raleigh, NC). She received her B.S. in Mathematics from Duquesne University (Pittsburgh, PA) and her Ph.D. in Applied Mathematics from Case Western Reserve University (Cleveland, OH). Her research is in the field of inverse problems and uncertainty quantification, approached from the Bayesian statistical perspective. She is specifically interested in the development of nonlinear Bayesian filtering algorithms for parameter estimation and their application to real-world problems relating to the life sciences.
