Clustering Spatio-Temporal Waves of Covid-19

It is commonly recognized that Covid-19 fluctuates in waves of infection activity, but much less is known about how waves differ or coincide when they are observed from different geographical locations. In this paper we aim to study the geographical patterns of infection waves throughout the Covid-19 pandemic. We propose a novel methodology for doing so, which both segments infection time-series data into waves and then clusters them together. With it we study US state and county level data, as well as data from European countries. From a clustering, we aim to bring understanding to viral spread by measuring geographical proximity and (for states and countries) similarity in terms of public health response.