Characterising COVID-19 Morbidity Dynamic Pattern across Neighbourhoods in New York City

This research implements partitioning around medoids (PAM) clustering algorithm on the dynamic time warping (DTW) distance calculated from the weekly variation of COVID-19 testing positive rate, aiming to profile and understand the dynamics of viral morbidity among neighbourhoods in New York City (NYC). With the specific focus on the residential usage neighbourhood tabulation areas (NTAs) in NYC, we classify 197 NTAs into six clusters characterising different COVID-19 dynamics. Based on their salient features, we interpret each cluster and anticipate possible future directions of this work.