A layman's note on a class of frequentist hypothesis testing problems

TL;DR

This paper discusses a framework for hypothesis testing where the costs of errors are quantified, advocating for optimization-driven test size selection to improve decision-making accuracy.

Contribution

It introduces a method for choosing test sizes based on error costs in simple hypothesis testing, enhancing traditional approaches.

Findings

Optimizing test size reduces overall error costs.

Quantifying error costs improves test decision strategies.

Adaptive test size selection outperforms fixed-size tests.

Abstract

It is observed that for testing between simple hypotheses where the cost of Type I and Type II errors can be quantified, it is better to let the optimization choose the test size.