It's ordered by current, 7 day MA of new cases per capita. If you want to compare where states are right meow, you need to compare per cap and in the same time slice. Rank ordering by peak might show something that you really do need independent evidence of: a spatial/soc trend
-
-
Replying to @Mihoda @DanielleFong
A set of data always *has* a rank ordering by peak, and plotting it can make it look meaningful, but it's not evidence of the spatial trend that it implies. Very dangerous to test that hypothesis. It's better to start out with the geo/social model and plot that ordering to check
1 reply 0 retweets 2 likes -
-
Replying to @DanielleFong
Just for you... I changed two lines of code. Peak... rank ordered by date of peak and size of peak.pic.twitter.com/rCX61jltu1
2 replies 0 retweets 1 like -
Replying to @Mihoda @DanielleFong
Give me 10 stock tickers and a pre-crash date and I'll whip up a plot over my lunch hour showing you the optical illusion of ordering by peak.
1 reply 0 retweets 1 like -
Replying to @Mihoda
should be some kind of ordering that's quasigeographical or includes network distance somehow.
1 reply 0 retweets 0 likes -
-
-
Replying to @DanielleFong
But I think we already fell into the pit. We're both talking about interstate transmission *because of* the plot. The predominant factor is almost certainly behavior driving regional R values. And currently, there is a t(R)end in the recent peakers if you get my drift.pic.twitter.com/FBJiHFAmNL
2 replies 0 retweets 1 like -
Replying to @Mihoda @DanielleFong
Regional R is intrastate transmission... and guess what kinds of behaviors are politicized now? https://www.forbes.com/sites/tommybeer/2020/06/18/nebraska-governor-says-local-governments-wont-receive-coronavirus-funding-if-they-require-masks/#4c519efa1c79 …
1 reply 0 retweets 1 like
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.


