Glen - Very interesting view on these three counties... numbers normalized to population count and population density are a "good start"!
The question of what a good "mixing model" is for different geopolitical demographics is fascinating. It seems like McKinley/Gallup is on one end of the spectrum (very low population density overall, but a strong concentration in *one* location (or small set of service/shopping locations IN Gallup) serving the whole county population) vs Bernalillo which has dozens of sub-communities where their sub-populations may stay "close to home" if not always "at home", shopping at one or two of their own neighborhood supermarkets/hardwares. Poverty (and rurality) may also correlate positively with delays in diagnosis. More people may be more used to just staying home and weathering out an illness since going to the doc or urgent care can have a significant hurdle financially and logistically (a tank of gas, requiring the only reliable family vehicle, etc). I'm assuming that the "diagnosed case" date is not the presumed date of exposure/contraction/onset-of-symptoms but rather the date of the return of a test or of a declaration of a health-care worker. - Steve > 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. > > > .-. .- -. -.. --- -- -..-. -.. --- - ... -..-. .- -. -.. -..-. -.. .- ... > .... . ... > FRIAM Applied Complexity Group listserv > Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam > unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com > archives: http://friam.471366.n2.nabble.com/ > FRIAM-COMIC http://friam-comic.blogspot.com/
.-. .- -. -.. --- -- -..-. -.. --- - ... -..-. .- -. -.. -..-. -.. .- ... .... . ... FRIAM Applied Complexity Group listserv Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com archives: http://friam.471366.n2.nabble.com/ FRIAM-COMIC http://friam-comic.blogspot.com/
