项目名称: 高维数据下多样本均值检验问题的研究
项目编号: No.11526070
项目类型: 专项基金项目
立项/批准年度: 2016
项目学科: 数理科学和化学
项目作者: 曹明响
作者单位: 合肥师范学院
项目金额: 2.5万元
中文摘要: 本项目拟从以下两个方面进行研究: (1)高维数据下的MANOVA中的检验问题。基于多样本均值的离差平方和,将已有文献中的两样本检验统计量进行两种不同形式的推广,构造了条件更弱、功效较高、适用范围更广的新的检验统计量。 (2)高维数据下多样本均值的线性假设的检验问题。首先将Bennett变换推广到高维的情形,然后基于新的Bennett变换和已有文献中构造两样本检验统计量的方法构造新的检验统计量。 在这个大数据时代,分析复杂数据和高维数据将是统计学的一个主要任务,所以,本项目具有重要的科学意义和应用价值。
中文关键词: 高维数据;多样本;多元方差分析;线性假设;
英文摘要: This project intends to research from the following two aspects: (1)Tests on MANOVA in high dimensional data. On the basis of sum of squares of deviations of sample means, the two sample tests in literature are extended to two different expressions; Then, new tests are proposed, which has weaker conditions, higher powers and broader applications than existed tests. (2) Linear hypothesis on sample means in high dimensional data. At first, the Bennett's transformation is extended to the high dimensional case. Then using this new Bennett transformation and the existed methods on two-sample tests, new tests are proposed. The analysis of complex or high dimensional data is an urgent work for statistics, the project is important both in theory and practice at the age of big data.
英文关键词: High dimensional data;Several samples;MANOVA;Linear hypothesis;