On the Permutation Distribution of Independence Tests
نویسنده
چکیده
One of the most popular class of tests for independence between two random variables is the general class of rank statistics which are invariant under permutations. This class contains Spearman’s coefficient of rank correlation statistic, Fisher-Yates statistic, weighted Mann statistic and others. Under the null hypothesis of independence these test statistics have a permutation distribution that usually the normal asymptotic theory used to approximate the p-values for these tests. In this note we suggest using a saddlepoint approach that almost exact and need no extensive simulation calculations to calculate the p-value of such class of tests. Some key words: Independence tests; Linear rank test; Permutation distribution; Saddlepoint approximation.
منابع مشابه
I I . I I I I I I I I _ I on a Class of Permutation Tests for Stochastic Independence
asymptotically most powerful permutation (rank order) tests is offered for testing independence of two stochastic variates when the observable random variables correspond to a finite or countable set of contiguous cells having an underlying continuous distribution. The theory is illustrated by some examples.
متن کاملBootstrap and permutation tests of independence for point processes
Motivated by a neuroscience question about synchrony detection in spike train analysis, we deal with the independence testing problem for point processes. We introduce non-parametric test statistics, which are rescaled general U -statistics, whose corresponding critical values are constructed from bootstrap and randomization/permutation approaches, making as few assumptions as possible on the u...
متن کاملar X iv : 1 40 6 . 16 43 v 4 [ m at h . ST ] 2 7 M ay 2 01 5 Bootstrap and permutation tests of independence for point processes
Motivated by a neuroscience question about synchrony detection in spike train analysis, we deal with the independence testing problem for point processes. We introduce non-parametric test statistics, which are rescaled general U -statistics, whose corresponding critical values are constructed from bootstrap and randomization/permutation approaches, making as few assumptions as possible on the u...
متن کاملConditional independence testing based on a nearest-neighbor estimator of conditional mutual information
Conditional independence testing is a fundamental problem underlying causal discovery and a particularly challenging task in the presence of nonlinear and high-dimensional dependencies. Here a fully non-parametric test for continuous data based on conditional mutual information combined with a local permutation scheme is presented. Through a nearest neighbor approach, the test efficiently adapt...
متن کاملcoin: A Computational Framework for Conditional Inference
The coin package implements a unified approach for conditional inference procedures commonly known as permutation tests. The theoretical basis of design and implementation is the unified framework for permutation tests given by Strasser and Weber (1999). For a very flexible formulation of multivariate linear statistics, Strasser and Weber (1999) derived the conditional expectation and covarianc...
متن کامل