Title :
An upper-bound on second-order dependency
Author_Institution :
Comput. Sci. Dept., SUNY - Univ. at Albany, Albany, NY, USA
Abstract :
In this work, we study the upper bound of second order statistical correlation. We provide a condition for a random variable reaching the upper-bound, and an algorithm that transform any variable to have the maximum second order statistical correlation.
Keywords :
correlation methods; statistical analysis; maximum second order statistical correlation; random variable reaching; second-order dependency; upper bound; Covariance matrices; Matrix decomposition; Random variables; Symmetric matrices; Transforms; Upper bound; Vectors; Information theory; Second order dependency; Upper-bound;
Conference_Titel :
Signal and Information Processing (ChinaSIP), 2013 IEEE China Summit & International Conference on
Conference_Location :
Beijing
DOI :
10.1109/ChinaSIP.2013.6625288
