In this article, extension of a bivariate exponential distribution to a bivariate Weibull distribution is discussed. First, statistical properties of the bivariate Weibull distribution are explored. Then, likelihood inference based on complete as well as right-censored bivariate data are discussed. Moment-based estimates are also developed. Through extensive Monte Carlo simulations, performance of the point and interval estimates are evaluated, and are found to be satisfactory. A two-dimensional warranty model, and the application of the bivariate Weibull distribution in the warranty model is discussed. Two case studies based on real datasets are provided for illustrative purposes. Finally, it is concluded that the bivariate Weibull distribution is quite useful to model bivariate data, especially, from reliability studies.
CO Author: Katherive Davies; William Volterman
Journal: REVSTAT - Statistical Journal
Click here
