Asymptotic Supervised Predictive Classifiers under Partition Exchangeability

The convergence of simultaneous and marginal predictive classifiers under\npartition exchangeability in supervised classification is obtained. The result\nshows the asymptotic convergence of these classifiers under infinite amount of\ntraining or test data, such that after observing umpteen amount of data, the\ndifferences between these classifiers would be negligible. This is an important\nresult from the practical perspective as under the presence of sufficiently\nlarge amount of data, one can replace the simpler marginal classifier with\ncomputationally more expensive simultaneous one.\n