Taxicab Correspondence Analysis and Taxicab Logratio Analysis: A Comparison on Contingency Tables and Compositional Data

In this paper, we attempt to see further by relating theory with practice: First, review the principles on which three interrelated well developed methods for analysis and visualization of contingency tables compositional data are erected: Correspondence based Benzécri’s principle distributional equivalence, Goodman’s RC association model Yule’s scale invariance, Aitchison’s subcompositional coherence. Second, introduce a novel index named intrinsic measure quality signs residuals choice method. The criterion is taxicab singular value decomposition, package TaxicabCA in R developed. We present minimal script thatcan be executed obtain numerical results maps paper. Third, flexible method constant added zero counts so that logratio can applied.