LGAug 6, 2017
Universally consistent predictive distributions
arXiv:1708.01902v24 citations
AI Analysis
This work addresses the challenge of reliable predictive distributions for statisticians and machine learning practitioners, though it appears incremental as it builds on existing IID frameworks.
The paper tackles the problem of probability forecasting under IID assumptions, introducing procedures that are universally consistent and satisfy small-sample validity.
This paper describes simple universally consistent procedures of probability forecasting that satisfy a natural property of small-sample validity, under the assumption that the observations are produced independently in the IID fashion.