Modelling the COVID-19 epidemics in Brasil: Parametric identification and public health measures influence

Renato Machado Cotta; Carolina Palma Naveira-Cotta; pierre magal

Here, as mentioned in the introduction, we first review an analytical parametric identification described in more details in [4] [5] [6] [7] , that from the initial phases of the epidemic evolution allows to explicitly obtain the unknown initial conditions of the model, while offering a reliable estimate for the transmission rate at the onset of the epidemy. Nevertheless, even after these estimates, a few other parameters in the model remain uncertain, either due to the specific characteristics of the physical conditions or reaction to the epidemy in each specific region, or due to lack of epidemiological information on the disease itself. Therefore, an inverse problem analysis was undertaken aimed at estimating the main parameters involved in the model, as summarized in Table 1 below. First, for the dataset on the accumulated reported cases for China, the focus is on the parametrized time variation of the transmission rate ( 0 and ) and the fraction of asymptomatic infectious that become reported ( 0 ), in this case assumed constant, followed by an effort to refine the information on the average times (1/ν and 1/η) through All rights reserved. No reuse allowed without permission.