Study on probability plot correlation coefficient of the log-Weibull distribution
Tóm tắt
The log-Weibull distribution is a variant of the three-parameter Weibull distribution. The probability plot of a distribution model is desired since it can help to decide on whether the model is appropriate for fitting a given dataset and can provide the initial estimate of the model parameters. The decision on the appropriateness of a distribution is somehow subjective. This paper presents a probability plot of the log-Weibull distribution (LWPP). The distribution of the probability plot correlation coefficient is studied. From this distribution, a lower confidence limit is determined for determining whether the probability plot correlation coefficient derived from a given data set is large enough. The appropriateness and usefulness of this study are illustrated by two real-world examples.
Tài liệu tham khảo
Jiang R Y, Wang T J. Log-Weibull distribution as a lifetime distribution [C]//Proceedings of International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering (QR2MSE 2013). Chengdu, China: IEEE, 2013: 851–854.
Jiang R Y. A drawback and an improvement of the classical Weibull probability plot [J]. Reliability Engineering and System Safety, 2014, 126: 135–142.
Jiang R Y. A new bathtub curve model with a finite support [J]. Reliability Engineering and System Safety, 2013, 119: 44–51.
Murthy D N P, Xie M, Jiang R Y. Weibull models [M]. Hoboken: John Wiley & Sons, 2003.
Balakrishnan N, Beutner E, Cramer E. Exact two-sample nonparametric confidence, prediction, and tolerance intervals based on ordinary and progressively type-II right censored data [J]. Mathematics and Statistics, 2010, 19(1): 68–91.