A data-driven tool for tracking and predicting the course of COVID-19 epidemic as it evolves

Norden E Huang; Fangli Qiao; Ka-Kit Tung

When time is normalized by T , the derivative is given by The above results are obtained for the case of a single introduction into a region of 905 infected at t=0 and we solve for the subsequent development of the epidemic from 906 that single source. Consider now a large region consisting of a number of small 907 regions, and the "seeding" of the infected occurs at different times for different 908 regions. The large region could be China, and the first infection could be Wuhan, 909 Hubei and then the regions outside Hubei. Then we may have for the China as a 910 whole data for the newly infected a sum of several Gaussians staggered in time. As 911 long as the Gaussians are not separated so much that there are different peaks in the 912 combined data, the combined data can still be considered as Gaussian, as is the case 913 in the real data. However, the standard deviation σ of the combined Gaussian is 914 inevitably larger and is no longer given by b: 915