New estimation methods of Covid-19 cases and reproduction number with
using a dynamic linear model and adaptive Kalman filter for the USA :
New York , Arizona, Texas and Florida utilizing data between Marc and
July 28th, 2020
Abstract
In this study, cumulative and daily cases are estimated online using a
discrete-time dynamic linear model (DLM) and Adaptive Kalman Filter
(AKF) based on the total COVID-19 cases between Marc h-July 28, 2020 in
USA-Florida, USA-Texas, USA-Arizona, USA-New York. Employing the data
collected between Marc and July 28, 2020, it is showed that the
discrete-time DLM in conjunction with AKF provides a good analysis tool
for modeling the daily cases made using the in terms of mean square
error (MSE) and . After estimating the number of cumulative cases, the
daily case number estimate was calculated. After calculating the daily
case number estimate, the reproduction number estimate was obtained. The
method is online. Only the data on the last day is sufficient. The AKF
has never been considered for such an application. To the best of our
knowledge, the estimation of COVID-19 has not been studied with this
method.