Is there a detailed description of the raw RKI_COVID19.csv -- AnzahlFall 131 ?

[denis-bz]

Dear Dr Gehrcke,
Is there a detailed description anywhere of the raw RKI_COVID19.csv with columns
0 ObjectId
1 IdBundesland
2 Bundesland
3 Landkreis
4 Altersgruppe
5 Geschlecht
6 AnzahlFall
7 AnzahlTodesfall
...

In particular, what can
359084,9,Bayern,SK München,A15-A34,W,131,0,2020/11/19 00:00:00,09162,"27.11.2020, 00:00 Uhr",0,-9,2020/11/19 00:00:00,-9,0,0,Nicht übermittelt
mean -- 131 cases on that one line ? There are many > 1
Ncase: max 131 np.bincount [ 0 97838 25223 10871 ...

Thanks,
keep up the good work -- if you're ever near München, IOU a 🍺

[jgehrcke]

Hey!

Is there a detailed description anywhere

I don't think that the RKI ArcGIS systems / "feature servers" have thorough reference documentation w.r.t. to the data set(s) you're thinking of here.

By common sense, however, I do think that the 131 in the line you have shown corresponds to AnzahlFall. That's probably the number of positive test cases on that day (because I happen to "know" that cumulative values in the RKI systems are usually prefixed with the word "Summe").

So, that line probably means that on 27.11.2020 in SK München they had 131 new Covid-19 cases of women in the age group between 15 and 34.

keep up the good work -- if you're ever near München, IOU a beer

Thank you, that's nice to hear! :)

[denis-bz]

Hi Dr. Gehrcke, do AnzahlFall < 0 or AnzahlTodesfall < 0 mean NaN, and you drop such lines ? It’s usually but not always -9 Thanks, cheers — denis

…

On 29 Nov 2020, at 22:15, Dr. Jan-Philip Gehrcke ***@***.***> wrote: Hey! Is there a detailed description anywhere I don't think that the RKI ArcGIS systems / "feature servers" have thorough reference documentation w.r.t. to the data set(s) you're thinking of here. By common sense, however, I do think that the 131 in the line you have shown corresponds to AnzahlFall. That's probably the number of positive test cases on that day (because I happen to "know" that cumulative values in the RKI systems are usually prefixed with the word "Summe"). So, that line probably means that on 27.11.2020 in SK München they had 131 new Covid-19 cases of women in the age group between 15 and 34.

[jgehrcke]

sorry, life is tough right now. will try to get back to you soon but hope you've made progress in the meantime w/o me! :)

[denis-bz]

Hi Dr Gehrcke, is there a url to get RKI_COVID19.csv with curl, e.g. https://www.arcgis.com/… ? I see AG_RKI_SUMS_QUERY_BASE_URL in your code but no def Thanks cheers — denis

[jgehrcke]

Hey!

I see AG_RKI_SUMS_QUERY_BASE_URL in your code but no def

Also see #208 and in particular this quote:

You could start poking around at https://services7.arcgis.com/mOBPykOjAyBO2ZKk/ArcGIS/rest/services -- this system contains ArcGIS feature servers not only for Covid things, but also for Berufsfeuerwehren and THW, etc. :).

[jgehrcke]

do AnzahlFall < 0 or AnzahlTodesfall < 0 mean NaN, and you drop such lines ?

Depends on what you'd like to know :-).

In the Covid19_RKI_Sums feature server (which is what I decided to use for building up the RKI CSV files in this repository) there is the metric called SummeTodesfall which is a cumulative value that I have looked at rather carefully; to make sure I understand what it means and that it matches official diagrams and other data sources. That's what I have been using to read the time series for "cumulative covid 19 death count over time". I am then building the derivative (over time) from that time series manually, yielding the daily change. The AnzahlTodesfall metric is something like a daily change -- but I have never really looked at it carefully.