Comparitive Evaluation of Epidemic-Type Point Process Models

Frederic P. Schoenberg

Published March 1, 2022, SCEC Contribution #11812

Several different frameworks have been proposed for modeling the spread of Covid-19, including compartmental models such as the SEIR (Susceptible → Exposed → Infectious → Removed) differential equation model, and branching point process models such as the Hawkes point process model. These two approaches for forecasting Covid-19 will be described briefly, and we will discuss how they have fared in predicting Covid-19 case counts and hospital demand especially from March 2020 to Spring 2021. In particular, the physical plausibility of the SEIR model is weighed against the parsimony and flexibility of the Hawkes and recursive models. The mathematical connection between Hawkes and SEIR models will be described. Relative to Hawkes models, SEIR models and their variants have been used far more widely to describe the Covid-19 pandemic as well as other infectious diseases such as Ebola and SARS. However, recent studies have suggested that Hawkes and recursive models may be more accurate. We will discuss why. We will also discuss what it has been like trying to estimate the spread of Covid-19 in real time, with particular focus on the work of the Los Angeles County Department of Health Services Task Force. We will also discuss results from other epidemics, including Ebola in West Africa, Coccidioidomycosis in California, SARS in China, Pertussis in Nevada, and Rocky Mountain Spotted Fever in California. These trends might help inform us about the future of Covid-19 in California, particularly when it comes to the spread of new variants.

Schoenberg, F. P. (2022, 03). Comparitive Evaluation of Epidemic-Type Point Process Models. Oral Presentation at Michigan State Statistics Colloquium.