Oblique rotation in correspondence analysis:A step forward in the search for the simplest interpretation

摘要

Correspondence analysis (CA) is a popular method that can be used to analyserelationships between categorical variables. It is closely related to several popularmultivariate analysis methods such as canonical correlation analysis and principalcomponent analysis. Like principal component analysis, CA solutions can be rotatedorthogonally as well as obliquely into a simple structure without affecting the totalamount of explained inertia. However, some specific aspects of CA prevent standardrotation procedures from being applied in a straightforward fashion. In particular, therole played by weights assigned to points and dimensions and the duality of CAsolutions are unique to CA. For orthogonal simple structure rotation, proceduresrecently have been proposed. In this paper, we construct oblique rotation methods forCA that take into account these specific difficulties. We illustrate the benefits of ouroblique rotation procedure by means of two illustrative examples.