Quote:
Originally Posted by Former DJ
Infectious Disease Experts Uncertain How Bad Coronavirus Will Get
This may be slow pony, but I came across this article on Nate Silver's 538 web site.
https://fivethirtyeight.com/features...to-get-either/
Two sobering outtakes from the article: (1.) The death rate will be (could be) anywhere from 200,000 to 1 million. (2.) There's a 73 percent probability of a "second wave" of the virus returning sometime in the fall.
The only qualifier in these predictions is a high degree of uncertainty. The "consensus" estimate could be way too high - or way too low.
That is not what the experts predicted. From the article "Experts’ estimates of the number of COVID-19 deaths in the U.S. in 2020 ranged from 4,000 all the way to 1 million, a huge range that highlights how much we still don’t know about this disease."
200K was an average prediction estimate, not a low end. And the confidence interval is extremely wide on that. They didn't give a 95% CI (probably as with this small sample size and wide variation, it wouldn't be any different than the bounds), but the 80% CI was also huge: "between 19,000 and 1.2 million."
The second wave prediction above does reflect the article accurately. Also noteworthy is that these experts guestimate that as of March 15, "only 12 percent of infections in the U.S. had been reported." That estimate is averaged from very wide individual estimates too, though, with responses ranging from 5% to 40% of cases having been reported.
This is the problem with all the stats people keep throwing around. We just don't have the data yet. This is especially true of infection and mortality rates, as we're just not testing enough to know. We know that our stats are skewed, but we have no idea how skewed. Some people are using that to paint pictures of doom and gloom, and others to say that reaction is way overblown. What we should, instead, be painting is a picture of uncertainty and the need to test widely.