[FRIAM] What Is the Real Coronavirus Toll in Each State? - The New York Times

[uǝlƃ ☣]

Based on this discussion, I divided by population (per county, not a normalized amount like 100k) and land area (not including water area). The results are interesting. There was a report about a Gallup hospital having problems. So, I used McKinley county (NM) for comparison.

The raw slopes still (I think) do the best to show what's happening. Dividing by population biases the data to magnify the low population county. Dividing by area magnifies the smaller counties (Bernalillo: 3k km^2, Santa Fe: 5k km^2, McKinley: 14k km^2). Dividing by both produces the same "phenotype" as the simple Δ's, but squashes out the profile shapes (e.g. the slight sigmoid in the Bernalillo slope).

My standard mix with DeKalb, King, & Denver (and now Hall as well) shows even more interesting behavior, dividing out both population and area how Hall has caught up with Denver (a really bad sign since Denver County is very dense, mostly just the city of Denver and the airport, an order of magnitude denser than Hall). But i won't spam the list with this stuff anymore.

On 5/6/20 3:13 PM, uǝlƃ ☣ wrote:
> I think Δcases/m^2 would be interesting.


-- 
☣ uǝlƃ
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nm.png
Type: image/png
Size: 579625 bytes
Desc: not available
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20200508/b07c492f/attachment-0001.png>