A note on testing complete independence for high dimensional data

作者：Mao Guangyu^*

来源：Statistics & Probability Letters, 2015, 106: 82-85.

DOI：10.1016/j.spl.2015.07.001

• HDLSS
• High dimension
• Independence test
• Universal asymptotics

The high dimensional independence test in Schott (2005) assumes (m, n) -> infinity and m/n -> gamma is an element of (0, infinity), where m signifies the dimension and n denotes the sample size. This paper notes that without the restriction m/n -> gamma, the test is still effective provided that (m, n) -> infinity, or n is fixed but m -> infinity.

• 出版日期2015-11
• 单位北京交通大学