Suppose I have two independent samples. How can I test if the two samples have been drawn from the same distribution using a permutation test? The sample sizes are equal. The population distributions are unknown.

  • $\begingroup$ You need to start with what kinds of differences you want most to pick up. If you're just after any kind of difference in distribution, the usual two sample test statistics (such as the Kolmogorov-Smirnov, for example) might be a suitable statistic for a permutation test.. $\endgroup$ – Glen_b Jul 21 '15 at 9:40
  • $\begingroup$ I wrote my answer to the duplicate specifically to address future related questions like this one. $\endgroup$ – whuber Jul 21 '15 at 14:54