Estimating methods for quantile residual life regression model with censored length-biased data

Area of Science:

• Biostatistics
• Survival Analysis
• Statistical Modeling

Background:

• Limited research exists on quantile residual life regression models compared to standard quantile regression.
• Censored and length-biased data present unique challenges in survival analysis.

Purpose of the Study:

• To develop statistical inference methods for quantile residual life regression models.
• To address challenges posed by censored and length-biased data.
• To propose robust estimation procedures for regression parameters.

Main Methods:

• Utilizing martingale theory to develop novel estimating equations.
• Implementing a two-stage procedure for estimating regression parameters.
• Avoiding direct estimation of the censoring variable's survival function.

Main Results:

• The proposed estimators demonstrate uniform consistency and weak convergence.
• Simulation studies indicate the composite martingale method slightly outperforms the standard martingale method.
• The methods were successfully applied to the Channing House dataset.

Conclusions:

• The developed martingale-based methods provide effective statistical inference for quantile residual life regression with complex data.
• The composite martingale approach offers enhanced precision in survival data analysis.
• This research contributes to the advancement of survival analysis techniques.