Bayesian $k$-record analysis for the Lomax distribution using objective priors

Abstract

In this paper, we present new perspectives    of the parameters of a Lomax  model that measure the  relevance of different priors on the posteriors using upper $k$th records.  The importance of  this analysis is shown through the  establishment of  convenient rankings based on objective priors.   Among several possible priors, such as Jeffrey's, reference and maximal data information priors, we identify those priors that satisfy specific  convergence concepts. For illustration purposes, we measure the relevance of the priors within simulated data and real medical data of  cancer patients.