Answers are grounded in this paper and cite their sources.
We consider the joint visualization of two matrices which have common rows columns, for example multivariate data observed at two time points split accord-ing to a dichotomous variable. Methods of interest include components analysis for interval-scaled data, or correspondence for frequency data or ratio-scaled variables on commensurate. A simple result in matrix algebra shows that by setting up the in a particular block format, matrix sum and difference components be visualized. The case when we have more than two matrices is also and the methodology is applied to data from the International Survey Program.
No takes yet. Share an insight, caveat, or question.
Michael Greenacre (2001) studied this question.