Evolution of Contagion by COVID-19 in El Salvador Applying SIR-Dynamic Simulations with the Monte Carlo Method

Hernández, Kevin, Rafael, Gomez-Escoto, Raúl, Henríquez, Arce, Ismael and Andrés, Alexis (2020) Evolution of Contagion by COVID-19 in El Salvador Applying SIR-Dynamic Simulations with the Monte Carlo Method. Universidad de El Salvador.

Text
Evolution COVID-19 in El Salvador.pdf - Published Version
Available under License Creative Commons Attribution.
Download (529kB)

Abstract

The COVID-19 pandemic is at the present in full swing in El Salvador and experience in other countries forces us to make drastic public and health policy decisions to contain the disease. This report presents some estimates of the evolution of the disease under the conditions of social distancing and home quarantine ordered by the authorities. Estimates and projections provide evidence based on SIR-type mathematical models, applying the Monte Carlo method, to establish and evaluate the critical phases of the pandemic and its effects, so that more efficient measures and strategies can be re-evaluated to continue containing the entry into greater critical phases. The time-dependent SIR was used to calculate differents paramenters of pandemic in the country, using official data from March 18 up May 4. We found the recovery rate has a value of ˆα = (0.0658 ± 0.0267)1/t with t measured in days, while the transmision rate is βˆ = (0.108 ± 0.001)1/t, therefore, the basic reproduction number was calculated with the value of Rˆ0 = (3.18 ± 0.21). The time-dependent SIR also was used to calculate the projections for infected cases and recovered cases, however, we analyzed the implementation of the Monte Carlo method in the numerical solution of infected cases. The maximum peak of the contagion is calculated using the solutions for infected cases, with and without applying Monte Carlo method, and it predicts between 1,320 and 1,488 individuals in infected state, and projecting that the time window for the critical period of the epidemic will be between the first and second week of June, while it would be attenuating only in mid-August. The error analysis include the error by the parameters and the prediction error.

Item Type:

Other

Creators:

Creators	Email	ORCID
Hernández, Kevin	kevinhernandezbel@hotmail.com	UNSPECIFIED
Rafael, Gomez-Escoto	gomez.escoto@ues.edu.sv	UNSPECIFIED
Raúl, Henríquez	raul.henriquez@ues.edu.sv	UNSPECIFIED
Arce, Ismael	ismael.arcev@gmail.com	UNSPECIFIED
Andrés, Alexis	alexis.andres1210@gmail.com	UNSPECIFIED