How does a permutation test for association differ from a paired treatment test?
StatisticsLevel: IntermediatePermutation experiment design
How does a permutation test for association differ from a paired treatment test?
Hints
- Which relationship is the null destroying?
- Compare pairings with sample labels.
Answer
For association, rearrange pairings between variables while retaining each marginal sample. For paired treatment effects, keep pairings and swap sample assignments within pairs.
Source and licence: SciPy — resampling and Monte Carlo inference · BSD-3-Clause · answer: Quantoros editorial