Download PDFOpen PDF in browser

Predicting COVID-19 (Coronavirus Disease) Outbreak Dynamics Using SIR-based Models: Comparative Analysis of SIRD and Weibull-SIRD

EasyChair Preprint 4607

5 pagesDate: November 19, 2020

Abstract

SIR type models are built by a set of ordinary differential equations (ODE) which are strongly initial value dependant. To fit multiple biological data with SIR type equations requires fitting coefficients of these equations by an initial guess and applying optimization methods. These coefficients are also extremely initial value dependent. In vast publication of these types, we hardly see, among simple to highly complicated SIR type methods, that these methods presented more than maximum two set of biological data. We propose a novel method here which integrate an analytical solution of infectious population using Weibull distribution function into any SIR type models. The Weibull-SIRD method has easily fitted 4 set of COVID-19 biological data simultaneously. It is demonstrated that predictions of Weibull-SIRD method for susceptible, infected, recovered, and deceased populations from COVID-19 in Kuwait and UAE are superior compared with SIRD original ODE model. The proposed method here open doors for new deeper studying of biological dynamic systems with realistic trends of biological data than providing some complicated cumbersome mathematical methods with little insight into real physics of biological data.

Keyphrases: COVID-19, Outbreak, SIR model, SIRD model, Weibull, coronavirus disease, epidemiological model

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
@booklet{EasyChair:4607,
  author    = {Ahmad Sedaghat and Shahab S. Band and Amir Mosavi and László Nádai},
  title     = {Predicting COVID-19 (Coronavirus Disease) Outbreak Dynamics Using SIR-based Models: Comparative Analysis of SIRD and Weibull-SIRD},
  howpublished = {EasyChair Preprint 4607},
  year      = {EasyChair, 2020}}
Download PDFOpen PDF in browser