Majorization for \((0,1)\)-matrices (Q2332460)

Majorization is known as an order relation for vectors, but it has been extended to orders for matrices. There are different ways of doing such an extension. The authors present several sorts of majorization for matrices and compare them. This concept has its roots in a functional analytic approach to basic notions in statistics, in connection with the theory of comparision of statistical experiments. The theory deals with the information content in families of probability measures, as an abstract representation of statistical experiments. The authors study classical matrix majorization and focus on (0,1)-matrices. They find certain necessary and sufficient conditions for such matrices to be majorized and get an algorithm detecting majorization.