Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/113917
Title: On the evolution of the COVID-19 epidemiological parameters using only the series of deceased: a study of the Spanish outbreak using Genetic Algorithms
Authors: Acosta González, Eduardo 
Andrada Félix, Julián 
Fernández Rodríguez, Fernando 
UNESCO Clasification: 530202 Modelos econométricos
3212 Salud pública
Keywords: Covid-19
Genetic Algorithm
Predictive Modelling
Sird Model Estimation
Dynamics, et al
Issue Date: 2022
Journal: Mathematics and Computers in Simulation 
Abstract: We propose a methodology for estimating the evolution of the epidemiological parameters of a SIRD model (acronym of Susceptible, Infected, Recovered and Deceased individuals) which allows to evaluate the sanitary measures taken by the government, for the COVID-19 in the Spanish outbreak. In our methodology the only information required for estimating these parameters is the time series of deceased people; due to the number of asymptomatic people produced by the COVID-19, it is not possible to know the actual number of infected people at any given time. Therefore, among the different time series that quantify the pandemic we consider just the number of deceased people to minimize the square sum of errors. The time series of deaths considered runs from March to the end of September and is divided into four sub-periods reflecting the different isolation measures taken by the Spanish government. The parameters that we can estimate are the time from the beginning of the disease, the transmission rate, and the recovery rate; these last two ratios are estimated in each of the different sub-periods. In this way the model considered has 2x4+1=9 parameters that are estimated jointly over the whole period from the data of deceased. Given the complexity of the model, to estimate the parameters that minimize the square sum of errors, a Genetic Algorithm is used. Our methodology confirms the effectiveness of the sanitary measures taken by the Spanish government showing a dramatic reduction in the basic reproductive number R0 during confinement; also, a further increase in R0 after the end of the alarm state decreed by the government on June 21 was detected. Our results also point out that the Patient Zero in the COVID-19 Spanish outbreak emerged between the end of December and early January, at least four weeks before January 31st, that was the moment when the Spanish authorities reported the first positive case.
URI: http://hdl.handle.net/10553/113917
ISSN: 0378-4754
DOI: 10.1016/j.matcom.2022.02.007
Source: Mathematics and Computers in Simulation[ISSN 0378-4754],v. 197, p. 91-104, (Julio 2022)
Appears in Collections:Artículos
Adobe PDF (920,44 kB)
Show full item record

Google ScholarTM

Check

Altmetric


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.