Scholarlink

<p>In this paper, we consider partially linear varying coefficient models with missing covariates and measurement errors in high dimensional data. We use inverse probability weighting and the error correction factor to correct biases caused by missing covariates and measurement errors. Based on the varying coefficient functions of B-spline method, we propose the nonconvex penalty function for simultaneous variable selection and estimation and study the asymptotic properties of parametric and nonparametric functions. Finally, our proposed method is validated by Monte Carlo simulations and actual data analysis.</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

Variable selection and estimation for high dimensional partially linear varying coefficient models with missing data and measurement errors based on quantile regression

  • 期刊论文

  • 发表日期：

    2025年12月3日

查看全文
  • Bingjie Cui,
  • Mengmei Xi,
  • Xuejun Wang

摘要

In this paper, we consider partially linear varying coefficient models with missing covariates and measurement errors in high dimensional data. We use inverse probability weighting and the error correction factor to correct biases caused by missing covariates and measurement errors. Based on the varying coefficient functions of B-spline method, we propose the nonconvex penalty function for simultaneous variable selection and estimation and study the asymptotic properties of parametric and nonparametric functions. Finally, our proposed method is validated by Monte Carlo simulations and actual data analysis.