Keynote 1 Speaker: Charmaine Dean, U Waterloo
Title: Statistical approaches to the analysis of wastewater surveillance data
Chair: Joel Dubin, U Waterloo
Date: Thursday, August 3, 2023
Time: 9:00 – 10:00 am
Abstract: Measurement of viral signal in wastewater is considered a useful tool for monitoring the burden of COVID-19, especially during times of limited availability in testing. Research has shown that the increases in wastewater viral signal can provide an early warning for an increase in hospital admissions. However, the association between the wastewater viral signal and COVID-19 hospitalizations may not be linear or consistent over time. A clear understanding of the time-varying and nonlinear association between the wastewater viral signal and COVID-19 hospitalizations is necessary. This talk discusses some statistical approaches for the analysis of wastewater surveillance data. We examine the use of a distributed lag nonlinear model to study the nonlinear exposure-response delayed association of COVID-19 hospitalizations and SARS-CoV-2 wastewater viral signal, using data from Ottawa, Ontario, Canada. This analysis enhances our understanding of the association between wastewater viral signal and COVID-19 hospitalizations when the wastewater treatment plant and health region are one-to-one match. We also discuss a spatial-temporal analysis for the case where multiple wastewater treatment plants serve a region, and when one wastewater treatment plant serves multiple regions, using data from the Netherlands.
Keynote 2 Speaker: Eric Tchetgen Tchetgen, U Penn Statistics
Title: Universal Difference-in-Differences
Chair: James Rosenberger, NISS
Date: Friday, August 4, 2023
Time: 9:00 – 10:00 am
Abstract: Difference-in-differences (DiD) is a popular method to evaluate treatment effects of real world policy interventions. Several approaches have previously developed under alternative identifying assumptions in settings where pre- and post-treatment outcome measurements are available. However, these approaches suffer from several limitations, either (i) they only apply to continuous outcomes and the additive average treatment effect on the treated, or (ii) they are not invariant to monotone transformations of the outcome, or (iii) they assume the absence of unmeasured confounding given pre-treatment covariate and outcome measurements, or (iv) they lack semiparametric efficiency theory. In this paper, we develop a new framework for causal identification and inference in DiD settings that satisfies (i)-(iv), making it universally applicable, unlike existing DiD methods. Key to our framework is an odds ratio equi-confounding (OREC) assumption, which states that the generalized odds ratio relating treatment and treatment-free potential outcome is stable across pre- and post-treatment periods. Under the OREC assumption, we establish nonparametric identification for any potential treatment effect on the treated in view, which in principle would be identifiable under the stronger assumption of no unmeasured confounding. Moreover, we develop a consistent, asymptotically linear, and semiparametric efficient estimator of treatment effects on the treated by leveraging recent learning theory. We illustrate our framework with extensive simulation studies and two well-established real-world applications in labor economics and traffic safety evaluation.
This is joint work with Chan Park, PhD.
About the Speakers
Dr. Charmaine Dean (Ph.D., University of Waterloo) is Vice-President, Research and International at the University of Waterloo where she provides leadership in research and innovation, commercialization, and internationalization. Dr. Dean is also a Professor in the Department of Statistics and Actuarial Science at Waterloo. Her main research focus is in biomedical areas such as mapping disease and mortality rates, mixture models for disease mapping, and spatial smoothing methods for disease mapping. She also develops statistical methodology for application in environmental research including water quality monitoring, forest ecology, fire management, fire occurrence prediction, smoke exposure estimation from satellite imagery, and modeling of stream flow for flood analysis and predictions. Dr. Dean holds Fellowships with the Institute of Mathematical Statistics, the Fields Institute, the American Statistical Association, and others. She has served in leadership roles internationally and nationally, related to equity and inclusion, statistics and data science, research, and computing infrastructure. She is Chair of Council for the Natural Sciences and Engineering Research Council of Canada.
Eric J. Tchetgen Tchetgen is a distinguished academic figure in the fields of Biostatistics, Epidemiology, and Statistics. He currently holds the position of Professor of Biostatistics in Biostatistics and Epidemiology at the University of Pennsylvania, where he also serves as the Luddy Family President's Distinguished Professor and Professor of Statistics and Data Science. With a strong expertise in various statistical domains, Tchetgen Tchetgen has made significant contributions to the fields of semiparametric theory, nonparametric statistics, causal inference, missing data, and epidemiologic methods.
Throughout his career, Tchetgen Tchetgen has focused on developing statistical and epidemiologic methodologies that effectively utilize the information contained within scientific data while minimizing assumptions about the underlying data generating mechanisms. His research interests encompass semiparametric efficiency theory, causal inference, missing data problems, statistical genetics, and mixed model theory. As an esteemed professor, Tchetgen Tchetgen has dedicated himself not only to advancing statistical knowledge through his research but also to imparting his expertise to the next generation of scholars. He engages in teaching activities, sharing his insights and fostering a deep understanding of the subject matter among his students. Tchetgen Tchetgen's contributions to the field have garnered recognition and acclaim. His accomplishments have been acknowledged through various awards and honors, solidifying his position as a prominent figure in the statistical and epidemiologic communities.