Skip to Main content Skip to Navigation
Conference papers

Seleção de modelos epidemiológicos via análise de sensibilidade global

Abstract : This paper propose a methodology for epidemiological model selection by using Akaike information criteria bringing as novelty the construction of a likelihood function based in the results of a global sensitivity analysis through the Sobol's indices obtained by using polynomial chaos expansion. The main ideia is to incorporate of the information about the influence of the parameters on the response to select a more interesting model inside of a set of candidates. The strategy is applied to a set of compartmental models compatible with those used to analyze the recent COVID-19 pandemic, allowing to compare them without the presence of experimental data.
Complete list of metadata

Cited literature [26 references]  Display  Hide  Download
Contributor : Americo Cunha Jr Connect in order to contact the contributor
Submitted on : Friday, October 23, 2020 - 5:57:43 AM
Last modification on : Thursday, October 29, 2020 - 3:04:35 AM
Long-term archiving on: : Sunday, January 24, 2021 - 6:11:43 PM


Files produced by the author(s)


  • HAL Id : hal-02975811, version 1



Michel Tosin, Americo Cunha Jr, Flávio Coelho. Seleção de modelos epidemiológicos via análise de sensibilidade global. XXIII Congresso Brasileiro de Automática (CBA 2020), Nov 2020, Congresso Virtual, Brazil. ⟨hal-02975811⟩



Les métriques sont temporairement indisponibles