Statistical analysis of bivariate failure time data with Marshall-Olkin Weibull models

Area of Science:

• Biostatistics
• Survival Analysis
• Reliability Engineering

Background:

• Bivariate failure time data are common in medical research and engineering.
• The bivariate Weibull model is widely used but lacks general estimation procedures for right-censored data.

Purpose of the Study:

• To develop and evaluate general estimation procedures for the bivariate Weibull model with right-censored failure time data.
• To address the gap in existing methods for fitting this model to complex datasets.

Main Methods:

• Proposed two general estimation procedures: a graphical approach and a marginal approach.
• Conducted an extensive simulation study to assess the performance of the proposed methods.
• Applied the methods to an illustrative real-world example.

Main Results:

• The simulation study demonstrated that both proposed approaches perform well in practical scenarios.
• The graphical and marginal methods provide reliable parameter estimation for bivariate Weibull models.
• The methods are effective even with right-censored failure time data.

Conclusions:

• The developed graphical and marginal estimation procedures offer practical solutions for fitting bivariate Weibull models.
• These methods enhance the analysis of bivariate failure time data, particularly in medical and reliability studies.
• The findings provide valuable tools for researchers dealing with censored survival data.