基于随机F-矩阵的高维双样本协方差矩阵相等性检验

Test of Equality of Two-Sample High-Dimensional Covariance Matrices Based on Random F-Matrix

用高维随机矩阵理论,对高维双样本协方差矩阵相等性的检验给出一种新方法.结果表明,利用高维随机F-矩阵线性谱统计量的中心极限定理给出检验统计量的极限分布,不仅适用于高维数据,而且对于非正态的情形仍有效.

We gave a new method to test the equality of two-sample high-dimensional covariance matrices based on random matrix theory in a high-dimensional framework.The results show that using the central limit theorem for the linear spectral statistics of high-dimen sional random F-matrices,the limit distribution of the test statistics is given,which is not only suitable for high-dimensional data but also valid for non-normal cases.