Comparing COVID-19 growth rates

[iamlost]

As webdevs we thrive (or suffer) on statistics. So, here be some European/western COVID-19 growth rates, by country, compared to Italy (as first in group to exceed 100 cases and so further along in growth timeline):
* France, Germany, Switzerland, United States: on track.
* Sweden, United Kingdom: were on track but recently levelled and now below track.
* Spain: well above track.
* Canada: too early to be sure, either on track or possibly beginning to be below track.
Note: given ~5-day no symptoms incubation period next weeks worth of data is already fixed even though unknown. A lag time that many have difficulty keeping in mind.

[BeeDeeDubbleU]

UK Stats in detail:

[en.wikipedia.org...]

[lammert]

Add the Netherlands to be on, or above track. With France still needing to report the number of new cases and mortality for today, the Netherlands is currently on fourth place at worldometers when sorting the countries on the number of deaths in the last 24 hours.

The daily new cases rate is increasing steadily, even though the health services stopped testing health workers and those showing only mild symptoms. A week ago the new confirmed cases included all new people discovered with the virus. Now it only includes those whose symptoms are so severe that they need hospitalization. That's good for the statistics but not so good for the overall health situation.

[graeme_p]

The PHE page is very good for an overview: [arcgis.com...]

[lammert]

That PHE page needs a login/password

[not2easy]

That link has a # in it so it is broken by forum software. Try pasting it to your address bar and you do not need a log-in.
The page covers GB:

https://www.arcgis.com/apps/opsdashboard/index.html#/f94c3c90da5b4e9f9a0b19484dd4bb14

[tangor]

Two I use:

[epidemic-stats.com...]
[ncov2019.live...]

[lucy24]

I noticed yesterday that the US has jumped from 8th to 6th in total number of confirmed cases. Going purely by population, this is to be expected, but still not heartening. (India is way, way down the list, which is comforting, since they are even poorer than the US.) My morning list of websites now includes the continuously updated map [coronavirus.jhu.edu] at Johns Hopkins.

[engine]

Mod note: Let's stick to the key authority sites and avoid drops to the many, many sites popping up all over the place. Non-authority site links will be removed.

[lucy24]

:: insert boilerplate about citing .gov or .edu sites whenever possible ::

[tangor]

Also bear in mind that not all reports are yet complete ... some testing is still in early stages, any country, any locale...

Needs be said: Some areas might not want to give accurate reports for whatever reason(s).

Population ages also makes a difference in reporting mortality rates. In other words, keep checking all reliable sources for the latest information.

[iamlost]

I have been using the 2019 Novel Coronavirus COVID-19 (2019-nCoV) Data Repository by Johns Hopkins CSSE [github.com] displayed as it’s convenient for graphing national/provincial time series [github.com].

Note: tangor is correct in that the data may be skewed depending on how and to what effect jurisdictions test and report. However (1) one works within constraints with any data set and identifying same is part of best practice; (2) when comparing with an existing graph result (in my instance Italy) one can see initial proof of concept in many/most repeating the graph line and most critically see outliers whether above or below line and investigate what might be causes.
Note: Sweden’s below line flat graph may be caused by lack of testing and so underreported cases.