文章基本信息

标题：testforDEP: An R Package for Modern Distribution-free Tests and Visualization Tools for Independence
本地全文：下载
作者：Jeffrey C. Miecznikowski ; En-shuo Hsu ; Yanhua Chen 等
期刊名称：R News
印刷版ISSN：1609-3631
出版年度：2018
卷号：10
期号：2
页码：282-295
语种：English
出版社：The R Foundation for Statistical Computing
摘要：This article introduces testforDEP, a portmanteau R package implementing for the first time several modern tests and visualization tools for independence between two variables. While classical tests for independence are in the base R packages, there have been several recently developed tests for independence that are not available in R. This new package combines the classical tests including Pearson’s product moment correlation coefficient method, Kendall’s τ rank correlation coefficient method and Spearman’s ρ rank correlation coefficient method with modern tests consisting of an empirical likelihood based test, a density-based empirical likelihood ratio test, Kallenberg data driven test, maximal information coefficient test, Hoeffding’s independence test and the continuous analysis of variance test. For two input vectors of observations, the function testforDEP provides a common interface for each of the tests and returns test statistics, corresponding p values and bootstrap confidence intervals as output. The function AUK provides an interface to visualize Kendall plots and computes the area under the Kendall plot similar to computing the area under a receiver operating characteristic (ROC) curve.