Distribution of the inter- and intra- inertia in conditional MCA (Q2488414)

In conditional Multiple Correspondence Analysis dependencies in a set of categorical variables are investigated with elimination of influence of some covariate \(T\) (also categorical). The idea is to partition the data cloud in \(T\) sub-clouds, re-centre each sub-cloud on its own centre of gravity to and consider the intra groups dispersion \(I_W\) (inertia within groups) and inter groups dispersion \(I_B\) (inertia between groups). Conditional independence can be tested by \(I_W\) and \(I_B\). Results of simulations are presented.