superforecasting_-_the_art_and_science_of_prediction_-_philip_e._tetlock_and_dan_gardner
Differences
This shows you the differences between two versions of the page.
Both sides previous revisionPrevious revision | Next revisionBoth sides next revision | ||
superforecasting_-_the_art_and_science_of_prediction_-_philip_e._tetlock_and_dan_gardner [2016/12/18 12:47] – [Reference] hpsamios | superforecasting_-_the_art_and_science_of_prediction_-_philip_e._tetlock_and_dan_gardner [2016/12/18 12:48] – hpsamios | ||
---|---|---|---|
Line 18: | Line 18: | ||
These were funded by IARPA (the intelligence community’s equivalent of DARPA). After studying and identifying a group of superforecasters and their characteristics, | These were funded by IARPA (the intelligence community’s equivalent of DARPA). After studying and identifying a group of superforecasters and their characteristics, | ||
- | The mechanism used to understand whether a prediction is good or not (an idea new to me at least) is a Brier score. "This was a new idea to me. The math behind this system was developed by Glenn W. Brier in 1950, hence results are called Brier scores. In effect, Brier scores measure the distance between what you forecast and what actually happened. So Brier scores are like golf scores: lower is better. Perfection is 0. A hedged fifty-fifty call, or random guessing in the aggregate, will produce a Brier score of 0.5. A forecast that is wrong to the greatest possible extent—saying there is a 100% chance that something will happen and it doesn’t, every time — scores a disastrous 2.0, as far from The Truth as it is possible to get." | + | The mechanism used to understand whether a prediction is good or not (an idea new to me at least) is a Brier score. The math behind this system was developed by Glenn W. Brier in 1950, hence results are called Brier scores. In effect, Brier scores measure the distance between what you forecast and what actually happened. So Brier scores are like golf scores: lower is better. Perfection is 0. A hedged fifty-fifty call, or random guessing in the aggregate, will produce a Brier score of 0.5. A forecast that is wrong to the greatest possible extent—saying there is a 100% chance that something will happen and it doesn’t, every time — scores a disastrous 2.0, as far from The Truth as it is possible to get." |
If your aim is to improve your own ability to make predictions, | If your aim is to improve your own ability to make predictions, |
/home/hpsamios/hanssamios.com/dokuwiki/data/pages/superforecasting_-_the_art_and_science_of_prediction_-_philip_e._tetlock_and_dan_gardner.txt · Last modified: 2020/06/04 11:31 by hans