Cai, Tony; Liu, Weidong; Xia, Yin - In: Journal of the American Statistical Association 108 (2013) 501, pp. 265-277
In the high-dimensional setting, this article considers three interrelated problems: (a) testing the equality of two covariance matrices <inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="uasa_a_758041_o_ilm0001.gif"/> and <inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="uasa_a_758041_o_ilm0002.gif"/>; (b) recovering the support of <inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="uasa_a_758041_o_ilm0003.gif"/>; and (c) testing the equality of <inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="uasa_a_758041_o_ilm0004.gif"/> and <inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="uasa_a_758041_o_ilm0005.gif"/> row by row. We propose a new test for testing the hypothesis <italic>H</italic> <sub>0</sub>: <inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="uasa_a_758041_o_ilm0006.gif"/> and investigate...