Prediction and Explanation in Social Systems

tags

Predictive Analytics

"When evaluating claims about predictive accuracy, however, preference should be given to studies that use standardized benchmarks that have been agreed upon by the field or, alternatively, to confirmatory studies that preregister their predictions"

Notes

current practices for evaluating predictions must be better standardized

NOTER_PAGE: (1 . 0.25378151260504206)

theoretical limits to predictive accuracy in complex social systems must be better characterized

NOTER_PAGE: (1 . 0.2680672268907563)

predictive accuracy and interpretability must be recognized as complements, not substitutes

NOTER_PAGE: (1 . 0.29411764705882354)

historical separation of explanation and prediction in the social sciences

NOTER_PAGE: (1 . 0.37478991596638656)

methods from machine learning that evaluate performance largely in terms of predictive accuracy

NOTER_PAGE: (1 . 0.8789915966386556)

How predictable is human behavior?

NOTER_PAGE: (2 . 0.2705882352941177)

we require not only the relevant baseline comparison—that is, the best known performance—but also an understanding of the best possible performance

NOTER_PAGE: (2 . 0.5336134453781513)