Failure of Calibration is Typical
Statistics Theory
2013-06-21 v1 Machine Learning
Statistics Theory
Abstract
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.
Cite
@article{arxiv.1306.4943,
title = {Failure of Calibration is Typical},
author = {Gordon Belot},
journal= {arXiv preprint arXiv:1306.4943},
year = {2013}
}
Comments
Forthcoming in Statistics and Probability Letters