Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 182 - New Approaches in Statistical Modeling of Infectious Diseases
Type: Invited
Date/Time: Tuesday, August 10, 2021 : 1:30 PM to 3:20 PM
Sponsor: SSC (Statistical Society of Canada)
Abstract #314461
Title: Time Series Approaches to Compare Covid-19 Mortality in the Province of Ontario, Canada, Across Three Epidemic Waves
Author(s): Charmaine Dean* and Georges Bucyibaruta and Elizabeth Renouf and Dexen Xi
Companies: University of Waterloo and University of Waterloo and University of Waterloo and National Research Council Canada
Keywords: Covid-19; Infectious Disease Resurgence; Logistic Growth Model; Time series; co-integration
Abstract:

The province of Ontario, Canada has experienced three epidemic waves of Covid-19. Hospitalizations and deaths, though lagging indicators, are considered an important metric for comparison of waves. For the purposes of understanding the trend in hospitalizations and mortality we utilize time series approaches for modeling outcomes. Cointegration analysis is employed to identify the long-run relationship between these processes. The outcomes are modeled through a shared latent stochastic error term in a novel framework that allows us to study the underlying correlation between two time series processes. We also develop a logistic growth model for the cumulative number of deaths through each wave. Although an empirical model, it incorporates conceptual elements that support the framework required for modeling any infectious disease where hospitalizations are required in management of the disease. By nature, the logistic growth model is deterministic, and so we induce stochasticity through incorporating the variability that is observed in modeling the daily counts and propose a tool that that can be used to quantify the behavior of the disease within a short time period.


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

Back to the full JSM 2021 program