Conformal e-testing

TL;DR

This paper introduces conformal e-testing, a method based on conformal e-prediction, which offers a new approach for hypothesis testing that inherits strengths from conformal prediction and has unique advantages.

Contribution

The paper systematically develops conformal e-testing, highlighting its strengths, limitations, and how it differs from and improves upon traditional conformal testing methods.

Findings

Conformal e-testing inherits strengths of conformal prediction.

It offers advantages over traditional conformal testing.

The paper discusses the limitations of conformal e-testing.

Abstract

There is a useful counterpart of conformal prediction for e-values, called conformal e-prediction. Conformal prediction can serve as basis for testing the assumption of exchangeability, leading to conformal testing. Similarly, conformal e-prediction can also serve as basis for testing. The resulting conformal e-testing looks very different from but inherits some strengths of conformal testing; it even has some advantages over conformal testing. In this paper we discuss systematically both strengths and limitations of conformal e-testing.