well done stats for the US and states but why no modeling for the World?
given that few sane rational people would take China at their word for fuckin anything how do the numbers add via this US only model that there are only 37,000 deaths worldwide?
yes China's Urn supply could be an indication but that means something like 25K really died in China. what are Italy's numbers out of "only" 37,000 so far?
if this projection pans out then what will the World stats be?
Originally Posted by The_Waco_Kid
I just finished reading the paper, which you can find here:
https://www.medrxiv.org/content/10.1...752v1.full.pdf
I don't have any kind of background in statistics, so didn't fully understand it, so take the following with a grain of salt.
One of the primary purposes was to forecast demands on hospitals in the state of Washington, and they expanded that to the USA on a state-by-state basis. So there was no mandate to do this for the world. Deborah Birx, who you see on stage with Trump all the time, referenced this study and said they're using it.
The bad news, from your point of view (not mine) is that this model used Wuhan data extensively. It was a curve fit of sorts, using data not only from Wuhan but also Italy, South Korea, the USA and other places. But since Wuhan is the only place where you've seen the death rate go up and then come way back down, it was emphasized. It also assumed that all states would implement four types of social distancing, within seven days of the date they published their paper: school closures, non-essential business closures including bars and restaurants, stay at home recommendations, and travel restrictions including closing down public transport.
Anyway, if you believe that the Chinese were massively understating deaths in Wuhan, then this model will massively underestimate deaths in the USA. It may also underestimate deaths for any states or localities that don't implement the four forms of social distancing. I think in Texas we've implemented all four.
So, I think in your view, garbage in and garbage out. I trust the Chinese numbers more than you do, although I think they went to measures that we won't, like quarantining all people they believe are carrying the virus, including the healthy ones, in facilities away from their families
Btw, their mean estimate was 82,000 deaths, but they had a range from 38,000 to 162,000 deaths in the USA with a 95% uncertainty interval. 162,000 deaths wouldn't be a bad outcome IMHO, although 82,000 sounds a lot better.
