Abstract

For many years, authors have been interested in developing methods for generating distributions that provide a flexible family to model lifetime variables. This paper proposes an exact inference approach to the family of proportional reversed hazard distributions based on the pivotal quantity, which yields exact confidence intervals with the shortest-length as well as reasonable estimators for the family of proportional reversed hazard distributions. In addition, the approach is extended to functions related to the unknown parameters using a generalized pivotal quantity such as the the goodness of fit test and entropy issues without computational complexity. The proposed method is illustrated through Monte Carlo simulations and real data analysis.

摘要

多年来，作者一直对开发生成分布的方法感兴趣，这些方法提供了一个灵活的族来对生命周期变量进行建模。本文提出了一种基于关键量的比例反向危险分布族的精确推理方法，该方法产生具有最短长度的精确置信区间以及比例反向危险分布族的合理估计量。此外，该方法使用广义关键量扩展到与未知参数相关的函数，例如拟合优度检验和熵问题，而无需计算复杂性。通过蒙特卡罗模拟和真实数据分析说明了所提出的方法。