Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 355 - Advanced Bayesian Topics (Part 4)
Type: Contributed
Date/Time: Thursday, August 12, 2021 : 10:00 AM to 11:50 AM
Sponsor: Section on Bayesian Statistical Science
Abstract #318976
Title: Bayesian Inference of COVID-19 Spread Dynamics in India
Author(s): Kai Yin*
Companies: Case Western Reserve University
Keywords: Bayesian Statistics ; Dynamic Model

We developed a modified SEIR model to study the transmission dynamics of COVID-19 in India. Traditional approaches in population-based models employ a uniform transmission rate; however, they are not suitable for the complexity of COVID-19. Here we take public mobility and time-dependent individuals' responses into account of the transmission rate, and further study how the individual behavior (e.g., social distance, mask-wearing, and personal hygiene) and governmental policies (e.g., national lockdown) would affect the transmission of COVID-19. A Bayesian method is applied to calibrate the developed model to publicly reported data (daily new tested positive cases, death, and recovery). The uncertainty of the parameters is naturally expressed as posterior distributions. The calibrated model was then used for estimating report ratios, scenario studies, and predictions for selected states. The numerical results showed the importance of government action and individual behavioral response to delay the early outbreak and control the outbreak size of COVID-19, which allows the public healthcare system to have enough time and capacity to respond before the availability of vaccines.

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

Back to the full JSM 2021 program