Skip to contents

This is a helper function. It takes an object created by 'bayesian_test' and spits intervals with evidence for either the null or alternative models.

Usage

interpret_bayes(bfs, evidence_thresh = 10, cluster_size = 10)

Arguments

bfs

An object returned by 'bayesian_test'.

evidence_thresh

Threshold to judge evidence. Defaults to 10. It is meant to be symmetrical: values above 10 indicate support for the alternative Hp, less than 1/10 for the null.

cluster_size

How many timepoints constitute a cluster?

Value

A list including intervals that support H0 or H1.