Efficient estimation of a linear transformation model for current status data via penalized splines

Area of Science:

• Statistics
• Biostatistics
• Survival Analysis

Background:

• Current status data presents unique challenges in survival analysis.
• Semi-parametric linear transformation models offer flexibility in modeling time-to-event data.
• Efficient estimation methods are crucial for practical application.

Purpose of the Study:

• To propose a flexible and computationally efficient penalized estimation method.
• To address challenges associated with current status data in semi-parametric models.
• To establish the theoretical properties and practical utility of the proposed method.

Main Methods:

• Utilized monotone B-splines for approximating the unknown monotone function.
• Developed a hybrid algorithm combining Fisher scoring and isotonic regression for efficient model fitting.
• Investigated asymptotic properties, including convergence rates and semi-parametric efficiency.

Main Results:

• Established asymptotic properties of penalized estimators, including optimal convergence rates.
• Demonstrated the semi-parametric efficiency of regression parameter estimators.
• Numerical experiments confirmed the finite-sample performance of the proposed method.

Conclusions:

• The proposed penalized estimation method is flexible and computationally efficient.
• The method provides a robust framework for analyzing current status data using semi-parametric models.
• The approach is validated through simulations and real-world data applications.