A nonparametric test is a hypothesis test that does not require the population distribution to be specified by a fixed, finite list of parameters. Many nonparametric tests are distribution-free under the null hypothesis, meaning that the null distribution of the test statistic is the same for a broad class of population distributions.
Nonparametric does not mean assumption-free. Depending on the test, exact validity can require conditions such as independent observations, continuity, symmetry, or invariance under specified permutations. Examples include the Kolmogorov-Smirnov test, permutation test, and rank test.
