I don't intend to scare anyone more than they already are by the pandemic, however, COVID-19 is scarier than I thought. I did some basic analysis on the number of people infected with the disease, and the results are stunning.
Unless France, Germany, Italy and countries hit by the virus don't shut down as China did (which is very unlikely), the total number of cases can be expected to reach 1 million people in each country by late March/early April. I have attached a Spreadsheet that outlines projections for COVID-19. I believe such projections are not public to maintain calm. People must have an understanding of what is being dealt with at this very moment.
The analysis is straightforward; I found data regarding the Total number of Coronavirus cases from here. The data was converted to the logarithm of total cases as the growth of the virus, when unconstrained, is exponential. Slopes and intercepts were obtained using Simple Linear Regression. Using these slopes and intercepts, I further extrapolated the logarithm of cases up to 1st April 2020. The logarithm of Total cases was then converted back to Total number of cases. While the dataset is time series and ideally one shouldn't use Linear regression for analysis, it gives a good idea about the growth of the virus.
Further information and observations can be found on the blog post. I shall continue updating the blog over the next few days. [I shall add facts and observations here in some time]
The spreadsheet I used can be found here. I aim to expand the number of countries analysed and include more relevant metrics like Total Active Cases in future iterations.
All suggestions and comments are welcome
Unless France, Germany, Italy and countries hit by the virus don't shut down as China did (which is very unlikely), the total number of cases can be expected to reach 1 million people in each country by late March/early April. I have attached a Spreadsheet that outlines projections for COVID-19. I believe such projections are not public to maintain calm. People must have an understanding of what is being dealt with at this very moment.
The analysis is straightforward; I found data regarding the Total number of Coronavirus cases from here. The data was converted to the logarithm of total cases as the growth of the virus, when unconstrained, is exponential. Slopes and intercepts were obtained using Simple Linear Regression. Using these slopes and intercepts, I further extrapolated the logarithm of cases up to 1st April 2020. The logarithm of Total cases was then converted back to Total number of cases. While the dataset is time series and ideally one shouldn't use Linear regression for analysis, it gives a good idea about the growth of the virus.
Further information and observations can be found on the blog post. I shall continue updating the blog over the next few days. [I shall add facts and observations here in some time]
The spreadsheet I used can be found here. I aim to expand the number of countries analysed and include more relevant metrics like Total Active Cases in future iterations.
All suggestions and comments are welcome
No comments:
Post a Comment