Satellite images and COVID-19: Predicting infections through nighttime lights

The effectiveness of the policies implemented during the COVID-19 pandemic is relevant both for assessing the impact caused and for deriving policy lessons in case such phenomena recur. The main objective here is to determine the relationship between human activity and reported COVID-19 cases. To measure human activity, a metric of nighttime lights is generated using satellite images. Conducting cointegration tests we found a long-term equilibrium relationship between the two variables, and through causality tests we confirmed that it is feasible to predict COVID-19 infections from changes in nighttime lights. Using econometric models for time series, it is shown that COVID-19 infections respond to changes in nighttime lights with high statistical significance and with a lag of up to two weeks, implying that the intensity of human activity that was happening now would have been useful for planning the resources that would be necessary two weeks later.