Accelerated failure time model under general biased sampling scheme.

Area of Science:

• Biostatistics
• Survival Analysis
• Epidemiological Methods

Background:

• Time-to-event data analysis often encounters biased sampling, particularly in cohort studies.
• Outcome-dependent sampling schemes can introduce complexities in survival time estimation.
• Existing methods may not adequately address general biased estimating schemes in semiparametric models.

Purpose of the Study:

• To propose a unified estimation method for semiparametric accelerated failure time (AFT) models.
• To address general biased estimating schemes in survival data analysis.
• To provide a robust statistical framework for analyzing complex cohort data.

Main Methods:

• Developed a bias-offsetting weighting scheme for regression covariate estimation.
• Utilized rank-based monotone estimating functions for regression parameters.
• Employed convex optimization for solving estimating equations.

Main Results:

• The proposed estimator is proven to be consistent and asymptotically normally distributed.
• Large sample properties of the estimator were derived.
• The method demonstrated effectiveness in simulations and real-world data applications.

Conclusions:

• The unified estimation method provides a robust solution for semiparametric AFT models under biased sampling.
• The approach is applicable to various sampling schemes, including length-bias and case-cohort designs.
• This work advances statistical methodologies for analyzing complex survival data in research settings.