Scholarlink

We study the problem of estimating circular densities when sample data are affected by measurement errors. We propose a deconvolution approach involving lower bias kernel estimators which take the additional source of bias due to the presence of measurement errors into account. Some asymptotic properties are discussed, and numerical results are provided.

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

Lower Bias Circular Density Estimation with Contaminated Data

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Marco Di Marzio,
  • Stefania Fensore,
  • Agnese Panzera,
  • Chiara Passamonti

摘要

We study the problem of estimating circular densities when sample data are affected by measurement errors. We propose a deconvolution approach involving lower bias kernel estimators which take the additional source of bias due to the presence of measurement errors into account. Some asymptotic properties are discussed, and numerical results are provided.