Computationally intensive statistical methods for the analysis of infectious disease outbreaks

In this thesis, we present two analyses of COVID-19 outbreaks in different contexts. Our first analysis consists of hypothetical outbreaks at Simon Fraser University and makes use of enrollment data from shortly before the start of the pandemic. We carry out a simulation study of possible outbreaks, exploring many regimes for disease severity, and also several levels of a control strategy. Our second analysis deals with real outbreaks in long-term healthcare facilities across the province of British Columbia. Some important features of these outbreaks are not available, so we use missing data methodology to fit our model. Before presenting our second analysis, we give a survey of computationally intensive methods for the analysis of missing data. This includes the methods used in our analysis, as well as many related tools which may be of interest for other, related problems.