Regression analysis of mixed recurrent-event and panel-count data with additive rate models

Area of Science:

• Statistics
• Biostatistics
• Epidemiology

Background:

• Event-history studies frequently utilize recurrent-event and panel-count data.
• Mixed event-history data, combining both types, presents unique analytical challenges.
• Existing literature on mixed data analysis is limited.

Purpose of the Study:

• To develop regression analysis methods for mixed event-history data.
• To address situations where subjects have both continuous and discrete observation periods.
• To provide robust statistical tools for analyzing complex event data.

Main Methods:

• Utilized the additive rate model for regression analysis.
• Developed estimating equation-based approaches for parameter estimation.
• Investigated finite sample and asymptotic properties of estimators.

Main Results:

• The proposed methodology provides effective estimation of regression parameters for mixed data.
• Numerical studies demonstrate the practical utility of the developed approach.
• The method was successfully applied to a Childhood Cancer Survivor Study.

Conclusions:

• The developed statistical framework offers a valuable tool for analyzing mixed event-history data.
• This methodology enhances the statistical analysis capabilities in fields like epidemiology and social sciences.
• The approach is robust and applicable to real-world research, as shown by the case study.