Inverse ensemble forecasting for COVID-19 outbreaks

We apply an inverse ensemble forecasting approach to COVID-19 outbreaks in a collection of U.S. counties containing college towns. Modelling disease dynamics with an SIR model, we define a time-dependent map from the infection parameters to infection levels at a specified time. We assume an unobserved probability distribution on the parameter space induces an output distribution on the infection levels. We compute a distribution on the parameter space through the formulation of a Stochastic Inverse Problem solved using disintegration of measures. This solution corresponds to a distribution over the possible infection curves which can be used to forecast future infection levels in an ensemble forecasting framework. We verify the method through a simulation study, then apply the method to experimental data. Results suggest the method can provide accurate forecasts under certain population and modelling assumptions, but that the SIR model does not adequately describe the disease dynamics in the COVID-19 outbreak in the population.