Semiparametric varying-coefficient model for interval censored data with a cured proportion.

Area of Science:

• Statistics
• Biostatistics
• Medical Data Analysis

Background:

• Varying-coefficient models are increasingly used in statistical research.
• Application to censored data analysis in medical studies is growing.
• Interval censored data with a cured proportion presents unique analytical challenges.

Purpose of the Study:

• To incorporate flexible semiparametric regression for interval censored survival data with a cured proportion.
• To develop a robust statistical framework for analyzing complex medical survival data.
• To address the need for advanced modeling techniques in medical research.

Main Methods:

• A two-part model was adopted to describe overall survival.
• Local polynomial regression with cross-validation was used for fitting unknown functional components.
• Bootstrap methods were proposed for statistical inference.

Main Results:

• Consistency and asymptotic distribution of the estimation were established.
• A BIC-type model selection method was developed for component specification.
• Extensive simulations demonstrated the performance of the proposed methods.

Conclusions:

• The developed methods provide a flexible and effective approach for analyzing interval censored survival data with cured proportions.
• The study offers valuable tools for statistical analysis in medical research, particularly for complex survival data.
• The application to decompression sickness data validates the practical utility of the methods.