Learn R Programming

USP (version 0.1.2)

U-Statistic Permutation Tests of Independence for all Data Types

Description

Implements various independence tests for discrete, continuous, and infinite-dimensional data. The tests are based on a U-statistic permutation test, the USP of Berrett, Kontoyiannis and Samworth (2020) , and shown to be minimax rate optimal in a wide range of settings. As the permutation principle is used, all tests have exact, non-asymptotic Type I error control at the nominal level.

Copy Link

Version

Install

install.packages('USP')

Monthly Downloads

182

Version

0.1.2

License

MIT + file LICENSE

Maintainer

Thomas Berrett

Last Published

January 27th, 2021

Functions in USP (0.1.2)

USPFourier

Independence test for continuous data
InfKern

Kernel for infinite-dimensional example
USPFourierAdapt

Adaptive permutation test of independence for continuous data.
FourierBasis

Fourier basis functions
DiscStat

Test statistic for dependence in contingency table
FourierKernel

Kernel matrix for Fourier basis
USP.test

Independence test for discrete data
KernStat

Test statistic calculated from two kernel matrices
USP

Permutation test of independence.
USPFunctional

Independence test for functional data
coeffs

Calculate coefficients of a function's series expansion
sumbasis

Kernel entries in infinite dimensional case