Abstract

LOPEZ PEZOA, Edgar  and  GRILLO, Sebastián Alberto. Analysis of the variability of reported COVID-19 over three weeks in relation to mobility parameters. Rep. cient. FACEN [online]. 2022, vol.13, n.1, pp.64-72. ISSN 2222-145X.  https://doi.org/10.18004/rcfacen.2022.13.1.64.

Determining the relationship of mobility patterns on the evolution of the COVID-19 pandemic has potential application in better policies to control the pandemic. Google makes available mobility data for each country where the average percentage changes of the population are recorded over 6 categories: 1) enclosed recreational spaces, 2) supermarkets and pharmacies, 3) transportation stations, 4) parks and open places, 5) workplaces and 6) residential areas. In addition, we have the series of daily reported COVID-19 cases throughout the pandemic. In this paper we apply a weekly smoothing to all time sequences and further calculate a variation factor between the number of reported cases in 21 days and the current number of reported cases. We then apply a (i) principal component analysis with the mobility series labeled according to the reported variation factor and (ii) a correlation analysis between the mobility series and the reported variation factor. The results on data from Argentina, Brazil, Canada, Chile, India, Mexico, Poland, Russia, Turkey, and the United States show that mobility data can be more or less determinant depending on the country. In addition, trends of movement in residential areas are significantly correlated to the reduction of reported cases in most countries, however the correlations of mobility variables may vary quite a bit from country to country.

Keywords : COVID-19; mobility; correlation analysis; principal component analysis; time series.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )