Link to the paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1621800 …
Show this thread
This paper shows that folks who successfully predict a big unexpected hit (like a VC who backs a company no one else thinks will win & turns out to be right) are often just really bad at predictions, since the way they reached their extreme results came from bad data analysis!pic.twitter.com/WS7OgVE6E0
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Or choose a different loss function. Squared error is not a good figure of merit for investments that have heavily compounded returns. The assumption of normality is another problem. A normal prior makes a predictor too conservative when predicting long-tailed samples.
- 1 more reply
New conversation -
-
-
"we were not able to test this explanation" then why in the literal fuck did anyone think it was a good idea to publish this?
-
Because, if you read the abstract of the paper (https://www.gwern.net/docs/predictions/2010-denrell.pdf …), the point is to document that empirically good predictors of extremes are just bad predictors in general.
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.