Fork of the project COVID19 Modeling version 2020-03-20 (Author: Joshua McGee, who forked the project FitVirus update of Milan Batista).
This version implements several features not available in the initial project:
* prefetch files
* several countries at once
* printing in PDF and PNG (in folders orgarnized by countries)
* reporting in text file
* error messages, more robus syntax
* variable dayforecast
* aggregation of regions, fix for France (Metropolitan area)
* online implementation on https://fitness.agroparistech.fr (section external)
* etc.

=======================================================================================================================
 COVID19Modelingv2 fits the last COVID-19 cases (updated daily) for one country, a selection of countries or all countries (169) and proposes forecasts
  FORECAST results are based on the assumption of logistic distribution, they are proposed "AS IS"
  The logistic distribution looks as the normal distribution in shape but has heavier tails (higher kurtosis), see https://en.wikipedia.org/wiki/Logistic_regression
  (when the dataset is too small or when the numbers of cases did not enter in the % exponential phase, the statistics is very poor)
 
  Syntax: COVID19Modeling(country,daysforecast)
    Default country = "selection", where selection is a list of 24 countries
    (selection = ["France" "Italy" "Spain" "Germany" "UK" "Belgium" "Portugal" "Croatia" "Denmark" "US" "Canada" "Australia" "China" "India" "Japan" "Korea, South" "Iran" "Russia" "Lebanon" "Netherlands" "Austria" "Denmark" "Turkey" "Israel"])
    Default number days of forecast = 14
    Examples:
        COVID19Modeling("France") or COVID19Modeling('France')
        COVID19Modeling(["France" "Italy"]) or COVID19Modeling({'France' 'Italy'})
        COVID19Modeling() or COVID19Modeling("selection") --> a selection of 24 countries
        COVID19Modeling("all") or COVID19Modeling('all') --> do the analysis for all countries in the list (169)

2020-03-20 - INRAe\Olivier Vitrac - rev. 2020-03-22

=======================================================================================================================


Project source: https://www.mathworks.com/matlabcentral/fileexchange/74632-covid19-modeling-fitvirus-update

The model was created by Milan Batista (fitVirus). The model is a data-driven model that fits epidemic data to a logistic curve. The goal of the model is to make local predictions about the viral spread and epidemic duration. The model can be used to provide accurate approximations in certain situations. "The regression convergence may fail for a pure initial guess or small data set. Therefore the method does not apply to the early stages of an epidemic. Also, results are useless if the regression statistic does not meet minimum criteria, say R^2 > 0.8, p-value < 0.05." (Milan Batista)

DISCLAIMER: Data and models are only indicative. Model will fail in certain situations. A rigorous statistical analysis should be performed on all results. The model fails when additional epidemic stages (not described by logistic function) are encountered. Use at your own discretion.

See for more info: https://www.researchgate.net/publication/339912313_Forecasting_of_final_COVID-19_epidemic_size_200320

Data is stored online and is provided via JHU CSSE from various sources including:
"The World Health Organization (WHO), DXY.cn. Pneumonia. 2020, BNO News,
National Health Commission of the People? Republic of China (NHC),
China CDC (CCDC), Hong Kong Department of Health, Macau Government, Taiwan CDC, US CDC, Government of Canada, Australia Government Department of Health, European Centre for Disease Prevention and Control (ECDC) and Ministry of Health Singapore