References & Citations
Mathematics > Statistics Theory
Title: High-Dimension, Low Sample Size Asymptotics of Canonical Correlation Analysis
(Submitted on 10 Sep 2016 (v1), last revised 13 Sep 2016 (this version, v2))
Abstract: An asymptotic behavior of canonical correlation analysis is studied when dimension d grows and the sample size n is fxed. In particular, we are interested in the conditions for which CCA works or fails in the HDLSS situation. This technical report investigates those conditions in a rather simplified setting where there exists one pair of directions in two sets of random variables with non-zero correlation between two sets of scores on them. Proofs and an extensive simulation study supports the findings.
Submission history
From: Sungwon Lee [view email][v1] Sat, 10 Sep 2016 02:06:58 GMT (139kb,D)
[v2] Tue, 13 Sep 2016 01:57:44 GMT (139kb,D)
Link back to: arXiv, form interface, contact.