Topics in multivariate nonparametric and robust methods

该项目将开发一些新的统计工具，用于分析医学研究，环境研究，图像和工业过程中产生的单变量，多变量以及非常高维的数据。这包括例如面部图像、笔迹、人类基因、万维网、社交网络、疾病流行、癌症筛查和预防试验、神经元和寿命数据等。用于分析这些数据的一些工具涉及通过使用数据深度的概念对多变量数据进行排序和分类。数据深度是高维数据集中给定高维点的中心性的度量。

This project will develop some novel statistical tools for the analysis of univariate, multivariate and also very high-dimensional data arising from medical research, environmental studies, images, and industrial processes. This includes for example images of faces, handwriting, human gene, world wide web, social networks, disease epidemics, cancer screening and prevention trials, neuronal, and lifetime data, etc. Some of the tools for analyzing such data involve ranking and sorting of multivariate data by using a concept of data depth. Data depth is a measure of centrality of a given high dimensional point within a high dimensional data set.