Single-index varying coefficient model for functional responses.

Area of Science:

• Statistical modeling
• Medical imaging analysis
• Neuroscience

Background:

• Large-scale functional imaging data are increasingly common in research.
• Correlating imaging data with clinical variables like age and gender is crucial for scientific discovery.
• Existing models may not fully capture complex associations in spatial functional data.

Purpose of the Study:

• To develop a novel single-index varying coefficient (SIVC) model.
• To establish a varying association between functional responses and covariates.
• To provide a robust statistical framework for analyzing spatially-referenced functional data.

Main Methods:

• Development of a single-index varying coefficient (SIVC) model.
• Estimation of varying coefficient functions, index function, and covariance function.
• Systematic integration of information across spatial grid points.
• Asymptotic property examination (consistency, convergence rate).

Main Results:

• The proposed SIVC model effectively estimates key components of the functional association.
• Simulation studies demonstrated the good finite-sample performance of the estimation procedure.
• Real data analysis on Alzheimer's Disease Neuroimaging Initiative (ADNI) white matter tract data confirmed SIVC model's superiority.

Conclusions:

• The SIVC model offers a powerful and accurate approach for analyzing functional imaging data.
• It provides advantages over traditional varying coefficient models in complex datasets.
• The methodology is validated through simulations and real-world neuroimaging application.