Scholarlink

Classification of data observed with errors is a relevant topic which has still not been fully explored. We address this problem in the directional context and discuss two nonparametric approaches which rely on kernel estimation of circular densities. We provide some asymptotic \(L_2\) properties for the proposed methods, along with some simulation results.

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

Circular Kernel Classification with Errors in Variables

  • 会议论文

  • 发表日期：

    2025年

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

摘要

Classification of data observed with errors is a relevant topic which has still not been fully explored. We address this problem in the directional context and discuss two nonparametric approaches which rely on kernel estimation of circular densities. We provide some asymptotic \(L_2\) properties for the proposed methods, along with some simulation results.