The material given below deserves further study. I now have data on 72 different countries in the form of time series. (The Jupyter Notebook is here). The time-series plots look diverse. Both individually and structurally.
They form the basis for a new phase in my project: from observation to simulation. I think it makes sense to try to further map not only the scientific forces of the virus, but also and concurrently the social forces that contribute to the willingness to accept and act on limitations, as well as the economic forces, and last but not least, the administrative and institutional forces involved.
The graphs linked to those countries are interesting and difficult to interpret at first glance. I need to investigate further. Perhaps they can play a role in calibrating toy worlds that mimic the effect of the mechanisms in their behavior. But first the overview.
Let me take a look at how I and presumably most of us understood the pandemic halfway through (what an optimistic suggestion, halfway through, of course I mean when it was on the road for 25 weeks – now we are at week 50). I have now adapted the program (it is here) so that I can display the results for an arbitrary period and for an arbitrary selection of jurisdictions. First I show two pictures, one for what we could see in week 25 (mid-June) and one for now, December 22 or week 52. In fig. 1, those images (for the Netherlands) are side by side.
The pictures in Fig. 1 show how difficult it is to deduce what will happen in the future from a picture about the past. If you only focus on the left graph, you only have rules of thumb and common sense to imagine what the future (in the right picture) will look like. This does not only apply after 25 weeks. This also applies after 50 weeks.
Maybe it helps to look at the rest of the world. After all, it is a pandemic. I give a number of overviews with country pictures. We can now say with some certainty that policies that influence people’s behavior are also reflected in the numbers. I give these numbers in a standardized form per country: the number of registered infections per ten thousand inhabitants in a jurisdiction and the number of deaths per million inhabitants in that jurisdiction.
The amount of pictures I show is overwhelming. They follow in 6 groups of 12. First from the American continents, then from Africa, and then from Europe, from the Near East (as viewed from the Netherlands), from the Far East and from a collection of maverick or remaining countries.
The groups start with a combined overview of the deaths per million inhabitants as they developed during the first 50 weeks of the pandemic. It shows how the countries can be compared with each other in protecting their populations against COVID deaths. The following plots show more detail about how the ratio of the numbers of registered infections and deaths per day per week develops per country. (The weekly numbers for deaths give the daily averages for that week.) And now for the overview.
For the American countries it looks like there everywhere there are more deaths per infection than 1 in a hundred.
With the exception of Paraguay, Uraguay and Venezuela (where the numbers are one or two orders lower), the number of deaths per million inhabitants in week 50 will be anywhere near nine hundred.
In all African countries, the total number of deaths per million inhabitants in week 50 is below 100.
With the exception of Uganda and Eritrea, as in the American countries, in all African countries the numbers of deaths are greater than one hundred times the number of infections (possibly the tide is turning in Nigeria and Uganda in this respect). Eritrea is a special case because, despite infections being recorded, there seem to be no COVID deaths at all there.
In Europe, the total numbers of deaths per million inhabitants in week 50 are fairly evenly spread over a 300-1600 range.
What is striking for the European countries is that they have all developed a first wave, that they have mastered it with measures, and that in the wake of this, the numbers were low in weeks 25-35. During that period, resistance to the measures seems to have arisen and grown. Subsequently, a second COVID-19 wave and political tensions were provoked, partly fueled by dissatisfaction and by “influencers.”
The Near East
Half of the countries that I have classified under the label “Levant +” have a death rate per million inhabitants in week 50 between (roughly) 200-600. They are Iran, Iraq, Oman, Morocco, Lebanon and Tunisia. In the other countries (Pakistan, Afghanistan, Syria, the Emirates, Egypt and Algeria) these numbers remain around 50.
In the Emirates, Oman and Lebanon, the numbers of registered infections are higher than 100 times the number of deaths for several weeks.
The far East
All countries in this group (with the exception of the Philippines and Myanmar) have a mortality rate per million below 50 in week 50. In all countries in this group, policies appear to be aimed at eradicating and controlling the virus.
In a few of these countries, control is under pressure: in Myanmar, Malaysia, the Philippines, Japan and South Korea, the daily numbers of deaths have been on the high side for a long time or show exponential growth, albeit at a (still) very low level.
Mistfits? Maverick countries?
In the group of misfits/maverick countries I have placed those countries that I could not put in a group. The pictures belonging to those countries are interesting and difficult to interpret at first glance. We need to investigate further. Possibly they can play a role in calibrating toy worlds that mimic the effect of the mechanisms in their behavior.
The material given above deserves further study. I think it makes sense to try to further map not only the scientific, biological forces of the virus, but also the social forces that contribute to the willingness to accept and act on limitations, as well as the economic forces and opportunities, and last but not least administrative and institutional forces.