Occam's razor for prior distributions with an application to null hypothesis significance testing

What is it about?

This paper provides a rationale for an interpretation of Occam's razor that transforms prior distributions into simpler prior distributions. The practical example is an application of the transformation to null hypothesis significance testing.

Featured Image

Why is it important?

The rationale for using Occam's razor to reduce the significance level from 0.05 to 0.001 is derived from first principles in this paper. The paper also generalizes that use of Occam's razor to incorporate different levels of caution toward the kinds of false positives that brought on the replication crisis.

Perspectives