Semiparametric Stochastic Modeling of the Rate Function in Longitudinal Studies

Area of Science:

• Biostatistics
• Biomedical Data Science
• Stochastic Modeling

Background:

• Longitudinal biomedical studies often require modeling biomarker profile changes.
• Understanding rate functions is crucial for tracking disease progression and treatment efficacy.
• Existing methods may have limitations in accurately forecasting future biomarker trajectories.

Purpose of the Study:

• To propose a novel semiparametric approach for modeling biomarker rate functions.
• To develop a method capable of forecasting future functional data.
• To provide a flexible modeling framework for longitudinal biomarker profiles.

Main Methods:

• Utilizing stochastic differential equations to define biomarker processes.
• Incorporating covariate dependence and parametric function variation.
• Developing an efficient Markov chain Monte Carlo algorithm for statistical inference.

Main Results:

• The proposed semiparametric method demonstrates competitive goodness-of-fit.
• The approach shows superior performance in forecasting future functional data compared to existing methods.
• The methodology is validated through simulation studies and application to real-world data.

Conclusions:

• The developed semiparametric approach offers a robust tool for analyzing longitudinal biomarker data.
• This method enhances the ability to predict future biomarker profiles, aiding clinical decision-making.
• The application to prostate-specific antigen profiles highlights its practical utility in biomedical research.