An empirical algorithm to forecast the evolution of the number of COVID-19 symptomatic patients after social distancing interventions

Abstract

We present an empirical algorithm to forecast the evolution of the number of COVID19 symptomatic patients in the early stages of the pandemic spread and after strict social distancing interventions. The algorithm is based on a low dimensional model for the variation of the exponential growth rate that decreases after the implementation of strict social distancing measures. From the observable data given by the number of tested positive, our model estimates the number of infected hindcast introducing in the model formulation the incubation time. We also use the model to follow the number of infected patients who later die using the registered number of deaths and the distribution time from infection to death. The relationship of the proposed model with the SIR models is studied. Model parameters fitting is done by minimizing a quadratic error between the data and the model forecast. An extended model is also proposed that allows a longer term forecast. An online implementation of the model is avalaible at www.ctim.es/covid19