Scholarlink

<p>Considering the nonlinear autoregressive model <i>x</i><sub><i>t</i></sub> = <i>f</i>(<i>x</i><sub><i>t</i>−1</sub>,…, <i>x</i><sub><i>t</i>−<i>p</i></sub>, <i>θ</i>)+<i>ϵ</i><sub><i>t</i></sub>, where <i>θ</i> is the <i>q</i>-dimensional unknown parameter, and <i>ϵ</i><sub><i>t</i></sub>s are the errors, we construct a composite quantile regression estimator of <i>θ</i>. Under some conditions, we obtain the strong consistency and asymptotic distribution of the proposed estimator. By simulation we show that the composite quantile regression estimator has a good performance.</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

Asymptotics of the composite quantile regression estimator for nonlinear autoregressive models

  • 期刊论文

  • 发表日期：

    2025年9月18日

查看全文
  • Ke-Ang Fu,
  • Yajuan Dong

摘要

Considering the nonlinear autoregressive model xt = f(xt−1,…, xt−p, θ)+ϵt, where θ is the q-dimensional unknown parameter, and ϵts are the errors, we construct a composite quantile regression estimator of θ. Under some conditions, we obtain the strong consistency and asymptotic distribution of the proposed estimator. By simulation we show that the composite quantile regression estimator has a good performance.