even a perfect predictor is going to have the real outcome fall outside their interval some of the time
-
Show this thread
-
if it's say 90%, that doesn't mean the answer should be in 90% of the experts' intervals. it means across many questions 90% of the answers should be in any expert's interval
1 reply 0 retweets 2 likesShow this thread -
(above where i said "perfect" i meant perfectly calibrated)
1 reply 0 retweets 1 likeShow this thread -
"It's astounding how few experts' confidence intervals included the correct estimate " is just wrong. some of the time it's perfectly reasonable for ALL expert confidence intervals to fail to include the correct estimate
3 replies 0 retweets 2 likesShow this thread -
like if i roll 99 on a d100 all the dice experts in the world will have been wrong when they predicted between 6 and 95. most of the time that won't happen but some of the time it will. it's not astounding
2 replies 0 retweets 4 likesShow this thread -
Replying to @VesselOfSpirit
Adam Strandberg Retweeted
i agree in principle but https://twitter.com/robertwiblin/status/1244970387596853254 …
Adam Strandberg added,
This Tweet is unavailable.2 replies 0 retweets 1 like -
Replying to @The_Lagrangian
"how can that not be in your confidence distribution?" it's easy. if you expect that growth will slow, continued exponential growth is outside the middle n% of your prediction
1 reply 0 retweets 0 likes -
Replying to @VesselOfSpirit @The_Lagrangian
the experts were wrong about this, but just because they were wrong about a thing, that doesn't mean they're insanely overconfident or whatever
1 reply 0 retweets 0 likes -
Replying to @VesselOfSpirit @The_Lagrangian
we can't even know HOW wrong they were without knowing how many % the confidence interval is. which nobody in that thread seems interested in
1 reply 0 retweets 0 likes -
Replying to @VesselOfSpirit @The_Lagrangian
and it's not like this was a question we randomly chose to look at. we're looking at it BECAUSE the experts were wrong on it
1 reply 0 retweets 0 likes
(in part)
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.