Was mobility a good predictor of new COVID cases in Havana’s first wave?

ABSTRACT
I will present an overview of the results of analyzing mobile phone data to create population mobility models in Cuba. In particular, I will present the impact of the NPIs limiting human mobility. We studied the Granger causality relation between the COVID cases time series in every Health Area in Havana and tried to predict the sequence in any given area based on the population mobility interaction between that area and other areas with cases. We show that at this level of detail, mobility is a negligible contribution to predicting new cases. However, extensive simulations on an e-Havana help to understand why it doesn’t work, and when it does work as a predictor.

BIO
Dr. Alejandro Lage Castellanos is the head of the Theoretical Physics Department and a full professor at Havana University, Cuba. He also leads the Complex System Research center in Cuba. He received his bachelor’s degree in Physics from Havana University Cuba in 2005, and his Ph.D. in 2012. His main research topics are Statistical Physics, Complex systems, Game theory, Human mobility models, and System biology. He has been working closely with the Cuban National Telecommunications Company and was part of the government advisor expert team during the COVID-19.