Полная версия этой страницы: average squared difference
average squared difference как лучше перевести этот показатель? разность средней квадратов - это корректно?
Цитата(Вале а @ 7.02.2012 - 17:52) 
average squared difference как лучше перевести этот показатель? разность средней квадратов - это корректно?
вариация?
Возможно точнее, среднеквадратическое смещение
http://polymer.bu.edu/ogaf/html/chp36.htm
http://polymer.bu.edu/ogaf/html/chp36.htm
спасибо. собственно вот:
The Brier score measures the total difference between the event (winning) and the forecast
probability of that event as an average squared difference. As a benchmark, a perfect forecaster would
have a Brier score of 0, a perfect misforecaster (predicts probability of win is 1 when loses and 0
when wins) would have a Brier score of 1, and a fence-sitter (forecasts every game as 50/50) would
have a Brier score of 0.25. Our pundit is doing reasonably well.
The Brier score measures the total difference between the event (winning) and the forecast
probability of that event as an average squared difference. As a benchmark, a perfect forecaster would
have a Brier score of 0, a perfect misforecaster (predicts probability of win is 1 when loses and 0
when wins) would have a Brier score of 1, and a fence-sitter (forecasts every game as 50/50) would
have a Brier score of 0.25. Our pundit is doing reasonably well.
Цитата(Вале а @ 7.02.2012 - 22:52)
спасибо. собственно вот:
The Brier score measures the total difference between the event (winning) and the forecast
probability of that event as an average squared difference. As a benchmark, a perfect forecaster would
have a Brier score of 0, a perfect misforecaster (predicts probability of win is 1 when loses and 0
when wins) would have a Brier score of 1, and a fence-sitter (forecasts every game as 50/50) would
have a Brier score of 0.25. Our pundit is doing reasonably well.
The Brier score measures the total difference between the event (winning) and the forecast
probability of that event as an average squared difference. As a benchmark, a perfect forecaster would
have a Brier score of 0, a perfect misforecaster (predicts probability of win is 1 when loses and 0
when wins) would have a Brier score of 1, and a fence-sitter (forecasts every game as 50/50) would
have a Brier score of 0.25. Our pundit is doing reasonably well.
Тогда: средний квадрат разницы величины и её оценки.
Цитата(p2004r @ 8.02.2012 - 13:15)
Тогда: средний квадрат разницы величины и её оценки.
да скорее всего так. спасибо.
Для просмотра полной версии этой страницы, пожалуйста, пройдите по ссылке.