We point out that the traditional notion of test statistic is too narrow, even for the purpose of conformal prediction. The most natural generalization of the traditional notion happens to be too wide. We propose another natural generalization which is arguably the widest reasonable generalization. The study is restricted to simple statistical hypotheses.
Original languageEnglish
Title of host publicationProceedings of Machine Learning Research
EditorsAlex Gammerman, Vladimir Vovk, Zhiyuan Luo, Evgueni Smirnov
Number of pages16
Publication statusPublished - 29 Aug 2019


  • exact p-value
  • randomized p-value
  • test statistic
  • lexicographic order

Cite this