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 -
Replying to @VesselOfSpirit
May be so, but then we’re severely over-provisioned in experts, beyond surge capacity in # of topics needing opinions.
1 reply 0 retweets 1 like -
-
Replying to @VesselOfSpirit
If-then. Why have experts concerned with prediction if they can’t try to be uncorrelated about it?
1 reply 0 retweets 0 likes -
-
Replying to @VesselOfSpirit @othercriteria
Vessel Of Spirit Retweeted Vessel Of Spirit
Vessel Of Spirit added,
1 reply 0 retweets 0 likes -
Replying to @VesselOfSpirit
If we’re asking experts for their opinion on what the two trailing digits of C*VID deaths are on, say, 2021-05-01, then sure.
1 reply 0 retweets 0 likes
my point is simply that some events are objectively surprising given finite information
-
-
Replying to @VesselOfSpirit
I think your argument has a reliance on monotonocity or ordering or implications holding across experts’ models.
2 replies 0 retweets 0 likes -
Replying to @othercriteria
i don't know what you mean by that. note i'm not saying this case is exactly like the d100, just that there's an element of d100ness in it
0 replies 0 retweets 0 likes
End of conversation
New conversation -
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.