Characterization of partially observed epidemics through Bayesian inference: application to COVID-19

Hybrid prediction of infections and deaths due to COVID-19 in two Colombian data series

The prediction of the number of infected and dead due to COVID-19 has challenged scientists and government bodies, prompting them to formulate public policies to control the virus' spread and public health emergency worldwide. In this sense, we propose a hybrid method that combines the SIRD mathematical model, whose parameters are estimated via Bayesian inference with a seasonal ARIMA model. Our approach considers that notifications of both, infections and deaths are realizations of a time series process, so that components such as non-stationarity, trend, autocorrelation and/or stochastic seasonal patterns, among others, must be taken into account in the fitting of any mathematical model. The method is applied to data from two Colombian cities, and as hypothesized, the prediction outperforms the obtained with the fit of only the SIRD model. In addition, a simulation study is presented to assess the quality of the estimators of SIRD model in the inverse problem solution.

Forecasting Multi-Wave Epidemics Through Bayesian Inference

We present a simple, near-real-time Bayesian method to infer and forecast a multiwave outbreak, and demonstrate it on the COVID-19 pandemic. The approach uses timely epidemiological data that has been widely available for COVID-19. It provides short-term forecasts of the outbreak's evolution, which can then be used for medical resource planning. The method postulates one- and multiwave infection models, which are convolved with the incubation-period distribution to yield competing disease models. The disease models' parameters are estimated via Markov chain Monte Carlo sampling and information-theoretic criteria are used to select between them for use in forecasting. The method is demonstrated on two- and three-wave COVID-19 outbreaks in California, New Mexico and Florida, as observed during Summer-Winter 2020. We find that the method is robust to noise, provides useful forecasts (along with uncertainty bounds) and that it reliably detected when the initial single-wave COVID-19 outbreaks transformed into successive surges as containment efforts in these states failed by the end of Spring 2020.