Data depth concept offers a variety of powerful and user friendly tools for robust exploration and inference for multivariate socio-economic phenomena. The offered techniques may be successfully used in cases of lack of our knowledge on parametric models generating data due to their nonparametric nature. This paper presents the R package DepthProc, which is available under GPL-2 licence on CRAN and R-forge servers for Windows, Linux and OS X platform. The package consist of among others successful implementations of several data depth techniques involving multivariate quantile-quantile plots, multivariate scatter estimators, local Wilcoxon tests for multivariate as well as for functional data, robust regressions. In order to show the package capabilities, real datasets concerning United Nations Fourth Millennium Goal and the Internet users activity are used.
翻译:数据深度概念提供了各种强大和方便用户的工具,用于对多种变式社会经济现象进行稳健的探索和推断,在由于非参数性质而缺乏关于生成数据的参数模型的知识的情况下,可以成功使用所提供的技术,本文件介绍了根据CRAN、R-Forge Windows、Linux和OS-X平台的RPL-2许可证提供的R包深度Proc。这套软件包括成功采用若干数据深度技术,包括多变式孔径-Quantile地块、多变式散射估计仪、多变式和功能数据的本地Wilcoxon测试以及功能数据。为了显示组合能力,使用了有关联合国第四个千年目标和因特网用户活动的真实数据集。