Copy the page URI to the clipboard
Gower, John C.
(2000).
DOI: https://doi.org/10.1016/S0167-9473(99)00050-X
Abstract
ten Berge (Comput. Statist. Data Anal. 24, 1997, 357–366) distinguished between rank-one and rank-two departures from symmetry. A re-examination suggests that there is no substantive difference between the two approaches unless the type of symmetry is constrained in some way. The relationship between rank and dimensionality in the context of asymmetry is clarified. Implications for diagnostic methods for distinguishing between different models of asymmetry are discussed, paying special attention to additive adjustments to a distance matrix.