吉林大学学报(理学版) ›› 2018, Vol. 56 ›› Issue (5): 1125-1135.

• 数学 • 上一篇    下一篇

基于Signature和Ⅱ型双删失系统寿命数据的统计推断

冯海林, 郭甜   

  1. 西安电子科技大学 数学与统计学院, 西安 710071
  • 收稿日期:2018-01-25 出版日期:2018-09-26 发布日期:2018-11-22
  • 通讯作者: 郭甜 E-mail:guotian9210@163.com

Statistical Inference Based on Lifetime Data of Signatureand Doubly TypeⅡ Censored  Systems

FENG Hailin, GUO Tian   

  1. School of Mathematics and Statistics, Xidian University, Xi’an 710071, China
  • Received:2018-01-25 Online:2018-09-26 Published:2018-11-22

摘要: 基于Signature理论和方法, 建立Ⅱ型双删失情形下元件寿命的统计推断模型, 用极大似然估计及其渐近理论, 推导出系统元件寿命分布参数的点估计及渐近置信区间估计的解析表达式, 并通过3种优化算法获得相关估计的数值解. 仿真结果表明, 基于Signature的似然函数统计推断模型对Ⅱ型双删失情形下元件寿命的推断有效.

关键词: 系统Signature, Ⅱ型双删失, 极大似然估计, 优化算法

Abstract: We established a statistical inference model of the component lifetime for the doubly typeⅡ censored cases  based on the theory and methods of Signature. By using the maximum likelihood estimation and its asymptotic theory, we derived the analytical expression for the point estimation and asymptotic confidence interval estimation of the component lifetime distribution parameters and  obtained the numerical solutions of the relevant estimation by three optimization algorithms. Simulation results show that the likelihood function statistical inference model based on Signature is effective for inference of the component lifetime under the doubly type Ⅱcensored cases.

Key words: system Signature, doubly typeⅡ censored, maximum likelihood estimation (MLE), optimization algorithm

中图分类号: 

  • O213.2