Scholarlink

<p>This paper provides a robust and efficient method for the varying-coefficient additive model with functional and longitudinal data and obtains a sparse estimate. A spline-based iterative algorithm is proposed for the proposed method. We provide asymptotic guarantees for the proposed method, showing that the proposed estimator achieves the <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(L^2\)</EquationSource> </InlineEquation> consistency. Simulation studies show that the proposed method outperforms other existing methods under the varying-coefficient additive model. We also apply the proposed method to the US high school dropout data and obtain that VCAM has the advantages of analyzing the interaction effect of time and covariates and extracting the time-varying effect of covariates on response.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A sparse estimate to varying-coefficient additive models for functional and longitudinal data

  • 期刊论文

  • 发表日期：

    2025年12月24日

查看全文
  • Yaxuan Zhao,
  • Zhimeng Sun,
  • Yuehan Yang

摘要

This paper provides a robust and efficient method for the varying-coefficient additive model with functional and longitudinal data and obtains a sparse estimate. A spline-based iterative algorithm is proposed for the proposed method. We provide asymptotic guarantees for the proposed method, showing that the proposed estimator achieves the \(L^2\) consistency. Simulation studies show that the proposed method outperforms other existing methods under the varying-coefficient additive model. We also apply the proposed method to the US high school dropout data and obtain that VCAM has the advantages of analyzing the interaction effect of time and covariates and extracting the time-varying effect of covariates on response.