Thursday, March 12, 2020

UPDATED: Forecasting The European Covid-19 situation on (a)live data

Background: Because I am co-chair of two upcoming events with international delegates end exhibitors, I had to put myself into the Corona situation when we had to decide if to move the events and how long ahead should we move them. I will not go into detail on the events themself since these are communicated through other channels. I have a background in working in very clean factories (Fabs) producing Logic and memory chips. I have been monitoring production processes with statistical process control methods and contamination control within and between Fabs as well as making risk assessments for contamination, process instabilities, and handling large amounts of deadly chemicals and gases safely. With this background, I was very surprised when I started looking into the published data on the Covid-19 outbreak - nowhere. I was not able to find graphs and plots that I could use to forecast and understand the situation. So I made my own. A while ago, I started to post these on Facebook, and people liked them and have given good feedback on how to improve them. This is the background why I now publish them here on my ALD blog, and I think I will keep on updating them. The data I get from open sources, and I will post the links later. Any feedback is most welcome! Stay tuned. Stay safe!

Please note - I am not an expert on viruses or epedemics. I do this because of the reasons mentioned above for my guidance. I may be wrong. If you want the original Excel-file please PM me (jonas.sundqvist@baldengineering.com)

Log scale plot of reported Covid-19 cases in interesting regions.

Same data set as above but normalized to inhabitants (reported cases per 1 million inhabitants)


Forecasting the European situation by extrapolation
Many experts have failed in forecasting the European situation. Here is my attempt to look at when will the European virus epicenters reach the same level (reported cases per million inhabitants) as the Huabei province in China where it all started.

I define that Europe has the following Epicentres today (2020-03-12):

  • Italy
  • Scandinavia
  • Germany
  • Spain
  • France

According to my extrapolations (see below), Scandinavia will have a Hubei situation (same number of reported cases per million inhabitants) 16th of March 2020 and Italy on the 19th of March 2020 if the current and upcoming actions have no positive effect. So bad news for Scandinavia is that the number of cases may go above the Italian situation. However, as for now the mortality rate in Schandinavia is very low as compared to Italy.

Spain looks like being on a similar trach whereas France and Germany are developing slower. The UK has the situation under some control and will not reach the Hubei-level in March.



To expand further I also looked today (12 March) on the US situation. As you can see below USA is reporting more and more cases for every day that goes and is set to cross the China line March 19 in one week from now.


1 comment:

  1. Clear graphs compared to others ive seen. Did you post links to sources somewhere yet?

    ReplyDelete