Failure of Calibration is Typical

Publication Year:
2013
Usage 264
Downloads 264
Social Media 1
Tweets 1
Repository URL:
http://philsci-archive.pitt.edu/id/eprint/9842
Author(s):
Belot, Gordon
preprint description
Schervish (1985b) showed that every forecasting system is noncalibrated for uncountably many data sequences that it might see. This result is strengthened here: from a topological point of view, failure of calibration is typical and calibration rare. Meanwhile, Bayesian forecasters are certain that they are calibrated---this invites worries about the connection between Bayesianism and rationality.