Statistical Analysis of Bivariate Failure Time Data with Type I and Type II Marshall–Olkin Weibull Models

Swar O. Ahmed, Khwazbeen S. Fatah, Sahib Esa

This paper discusses the problem of parametric analysis of a probability distribution known as type I and II Marshall-Olkin bivariate Weibull distribution, which is frequently used in the analysis of survival and reliability. It is remarkable that there is no specific technique for fitting the bivariate right-censored failure time data to type II Marshall-Olkin bivariate Weibull distribution. Although, there are few techniques for estimation parameter for type I MOWE, none of them could accurately estimate the additive parameter of Type II MOWE. Therefore, this paper presents a simple procedure to estimate the additive parameter of the new distribution dependent on type I MOWE; in this methodology the parameters of type I MOWE and Non-parametric bivariate survival function are estimated. Consequently, by using simple regression, the additive parameter is estimated. Furthermore, the proposed approaches are applied on bivariate right-censored survival time data to illustrate and assess the performance.

Volume 12 | 02-Special Issue

Pages: 253-264

DOI: 10.5373/JARDCS/V12SP2/SP20201068