Ordering properties of the largest order statistics from Kumaraswamy-G models under random shocks

Abstract

In this paper we compare the maximums of two independent and heterogeneous samples each following Kumaraswamy-G distribution with the same and the different parent distribution functions using the concept of matrix majorization. The comparisons are particularly carried out with respect to usual stochastic ordering when each sampling unit experiences a random shock. It is also shown that, under similar conditions no reversed hazard rate ordering exists between maximums.