Risk management for return enhancement

Derived from a century-old and largely forgotten theorem involving the basic characteristics of Pearson’s correlation, Mark Lundin and Steve Satchell introduce a non-linear and asymmetric dependence method that indicates that seeking high correlation in some regimes and negative correlation in others tends to be more profitable than holding assets that are uncorrelated at all times

Myriad dependence measures are actively implemented in financial market applications today. Pearson's productmoment correlation coefficient (as well as covariance) remains ubiquitous in practice, despite its inability to capture nonlinear dependence and its assumption of homoscedasticity.

CLICK HERE TO VIEW THE PDF

Rank correlation methods capture nonlinear dependence, though they do so by replacing sample magnitudes with ordinal ranking.