Estimation of the cumulative baseline hazard function for dependently right-censored failure time data

Area of Science:

• Survival analysis
• Statistical modeling
• Biostatistics

Background:

• Frailty models account for unobserved heterogeneity in survival data.
• Common frailty models are essential for analyzing multiple related failure times.
• Archimedean copulas provide a flexible framework for modeling dependencies.

Purpose of the Study:

• To investigate properties of bivariate frailty models with common frailty.
• To develop a novel estimator for cumulative baseline hazard functions.
• To introduce a graphical model checking procedure for these models.

Main Methods:

• Derivation of a formula for cumulative baseline hazard functions.
• Development of a new non-parametric estimator for bivariate frailty regression.
• Application of the estimator in a graphical model checking procedure.
• Fitting a leukemia dataset using the proposed bivariate frailty model.

Main Results:

• A useful formula for cumulative baseline hazard functions was established.
• A new estimator for cumulative baseline hazard functions in bivariate frailty regression was developed.
• A graphical model checking procedure based on the new estimator was presented.
• The proposed model was successfully fitted to leukemia data.

Conclusions:

• The developed methods provide valuable tools for analyzing bivariate survival data with common frailty.
• The new estimator and graphical procedure enhance the application of frailty models.
• The leukemia data analysis demonstrates the practical utility of the proposed approach.