Speaker
Abstract: Laboratories use group (pooled) testing to reduce the time and cost associated with screening large populations for infectious diseases. However, these benefits come at the expense of a potentially complex data structure which can hinder surveillance efforts. In this talk, I will give a survey of regression methods for group testing data. These methods encapsulate a number of primary models for the latent disease response(s), variable selection for fixed and random effects, and techniques to estimate assay sensitivity and specificity. Although this is a survey talk, I will also discuss some open problems in group testing and potentially new applications. I will highlight my involvement with the State Hygienic Laboratory at University of Iowa, the largest public health lab in Iowa, and how they use group testing to screen thousands of residents for chlamydia and gonorrhea each year.
Bio: Dr. Tebbs is Professor in the Department of Statistics at University of South Carolina. He obtained his BS in Mathematics and MS in Statistics from University of Iowa and his PhD in Statistics from North Carolina State University. His primary research interests are in the development of statistical methods for categorical data, especially aggregated or group tested data and their application in infectious disease testing, as well as in general biostatistical methods and problems involving ordering or shape restrictions. He is an elected member of the International Statistical Institute, a fellow of the American Statistical Association, and his research program is funded by the National Institutes of Health (NIH). He is a former member of the Biostatistical Methods and Research Design (BMRD) Study Section of the NIH, and he recently completed a three-year term as Editor of American Statistician.
Zoom Link: https://duke.zoom.us/j/93960612613?pwd=Dos6w77mZ8qDyQw1o22soeI4zeV0Fh.1