The permutation test, inference by reshuffling labels
You have two samples (say, returns from strategy A and strategy B) and want to test whether they come from the same distribution, without assuming normality.
Explain the permutation test: the algorithm, why it works, and where it can mislead.
Show a hint
Under the null hypothesis that the two groups are identical, the group labels carry no information. What can you do to a label that carries no information?
Your answer
This one is open-ended. Work it through, then check your reasoning against the full solution.