Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 284 - Statistical Learning for Dependent and Complex Data: New Directions and Innovation
Type: Invited
Date/Time: Wednesday, August 5, 2020 : 10:00 AM to 11:50 AM
Sponsor: Section on Statistics in Marketing
Abstract #309454
Title: Spatiotemporal Dynamics, Nowcasting and Forecasting COVID-19 in the United States
Author(s): Guannan Wang and Lily Wang* and Yueying Wang
Companies: College of William and Mary and Iowa State University and Iowa State University
Keywords: Trivariate splines; Functional principal component analysis; Image analysis; Semiparametric efficiency; Tetrahedral partition

Since December 2019, the outbreak of COVID-19 has spread globally within weeks. To efficiently combat COVID-19, it is crucial to have a better understanding of how far the virus will spread and how many lives it will claim. Scientific modeling is an essential tool to answer these questions and ultimately assist in disease prevention, policymaking, and resource allocation. We establish a state-of-art interface between classic mathematical and statistical models to investigate the dynamic pattern of the spread of the disease. We provide both short-term and long-term county-level prediction of the infected/death count for the US by accounting for the control measures, mobility and local features. Utilizing spatiotemporal analysis, our proposed model enhances the dynamics of the epidemiological mechanism, which helps to dissect the spatial and temporal structure of the spreading and predict how this outbreak may unfold through time and space in the future. To assess the uncertainty associated with the prediction, we develop a projection band based on the envelope of the bootstrap forecast paths. Our empirical studies demonstrate the superior performance of the proposed method.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program