A Copula Based Stress-Strength Reliability Estimation with Lindley Marginals

被引:3
作者
James, A. [1 ]
Chandra, N. [1 ]
Pandey, M. [2 ,3 ]
机构
[1] Pondicherry Univ, Ramanujan Sch Math Sci, Dept Stat, Pondicherry 605014, India
[2] Banaras Hindu Univ, Inst Sci, Dept Zool, Varanasi 221005, Uttar Pradesh, India
[3] Banaras Hindu Univ, Inst Sci, DST Ctr Math Sci, Varanasi 221005, Uttar Pradesh, India
来源
JOURNAL OF RELIABILITY AND STATISTICAL STUDIES | 2022年 / 15卷 / 01期
关键词
Stress -strength reliability; Lindley distribution; Fralie-Gumble; Morgenstern; maximum likelihood estimation; inference function margins; semi -parametric method; Monte -Carlo simulation; MODEL; DEPENDENCE;
D O I
10.13052/jrss0974-8024.15114
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
The stress-strength model is a basic tool used in evaluating the reliability (R). It shows that a component or system with stress (Y) and strength (X) will fail if the stress exceeds the strength, and its counterpart allows it to function. Usually, the statistical independence between X and Y are assumed and reliability models are extensively developed in the literature. However, in real life, there are many situations in which the dependence stress-strength is taken into account. So it is important to consider and model the asso-ciation between them. In this paper, we estimated R when the stress and strength parameters are linked by a Fralie-Gumble-Morgenstern copula with Lindley marginals. The estimates of reliability and dependence parameter are obtained by using maximum likelihood estimation (MLE), inference function margins (IFM), and semi parametric (SP) methods. In addition, the length of the asymptotic confidence interval and the coverage probability of the dependence parameter are also computed. A simulation study is performed to evaluate the effectiveness of the various estimates, and a real data set is also used for illustrative purposes.
引用
收藏
页码:341 / 379
页数:39
相关论文
共 54 条
[31]  
Kotz S., 2003, The Stress-Strength model and its Generalizations: theory and applications, DOI [10.1142/5015, DOI 10.1142/5015]
[32]   AIC and BIC - Comparisons of assumptions and performance [J].
Kuha, J .
SOCIOLOGICAL METHODS & RESEARCH, 2004, 33 (02) :188-229
[33]   Estimation of R=P[Y<X] for three-parameter generalized Rayleigh distribution [J].
Kundu, Debasis ;
Raqab, Mohammad Z. .
JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION, 2015, 85 (04) :725-739
[34]   SOME CONCEPTS OF DEPENDENCE [J].
LEHMANN, EL .
ANNALS OF MATHEMATICAL STATISTICS, 1966, 37 (05) :1137-&
[35]  
LINDLEY DV, 1958, J ROY STAT SOC B, V20, P102
[36]  
Lindley DV, 1965, Introduction to probability and statistics from a bayesian viewpoint-Part 2: inference
[37]   The Lindley distribution applied to competing risks lifetime data [J].
Mazucheli, Josmar ;
Achcar, Jorge A. .
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2011, 104 (02) :188-192
[38]   REGRESSION WITH FRAILTY IN SURVIVAL ANALYSIS [J].
MCGILCHRIST, CA ;
AISBETT, CW .
BIOMETRICS, 1991, 47 (02) :461-466
[39]  
Morgenstern D., 1956, Mitteilingsblatt fur Mathematische Statistik, V8, P234
[40]  
Nelsen R. B, 2006, Lecture Notes in Statistics