An efficient penalized estimation approach for semiparametric linear transformation models with interval-censored...

Area of Science:

• Statistics
• Biostatistics
• Semiparametric Models

Background:

• Interval-censored data presents challenges in statistical modeling.
• Flexible transformation models are crucial for analyzing complex relationships.
• Dimensionality reduction is key for efficient semiparametric model estimation.

Purpose of the Study:

• To develop an efficient estimation method for flexible transformation models with interval-censored data.
• To approximate unknown monotone functions using monotone B-splines for dimensionality reduction.
• To enable computationally efficient parameter estimation and statistical inference.

Main Methods:

• Approximation of monotone functions using monotone B-splines.
• Penalization techniques for computationally efficient estimation.
• Nested iterative Expectation-Maximization (EM) algorithm for model fitting.
• A novel variance-covariance estimation approach for large-sample inference.

Main Results:

• The proposed method achieves optimal convergence rates for the monotone function estimator.
• Regression parameter estimators are shown to be asymptotically normal and efficient.
• The penalized procedure demonstrates robust performance in numerical experiments.

Conclusions:

• The developed methodology provides an efficient and statistically sound approach for analyzing interval-censored data using flexible transformation models.
• The R package PenIC facilitates the implementation of this penalized estimation technique.
• The approach is validated through simulations and a real-world signal tandmobiel study.