This is an update as of April 15th, 2020.
We are analyzing the existing available data on daily deaths caused by the SARS-CoV2 virus and use that in conjunction with certain simple models to predict the evolution of the disease in certain geographical areas. You can read the full introduction here.
Today was a small uptick in the total number of deaths, but despite that, most of the regions are past the peak.
The new addition today is Ohio, it’s passed the peak as well.
Linear model fit is applied to see how external factors influence the outcome of the outbreak. The considered external factors were: temperature in the period around the beginning of the outbreak, dew point in the period of the beginning of the outbreak, population density of the region, population size of the region and percentage of people over 65.
The period around the beginning the outbreak is defined as 3 weeks before we record the first 3 deaths to 1 week after that point.
The size of the outbreak shows the stronger fit with an AICc of 593 and for the size the most predictive variable was the dewpoint with a Pvalue of 0.98. The size is anticorrelated with the dewpoint in the results of this model.
We will follow this dependency in the subsequent days and tomorrow will present all the results of the linear model fit.
Your feedback is important to us. Please send it to email@example.com