Scholarlink

We investigate change points in the autoregression parameters of AR(1) and RCA(1) sequences. The errors are allowed to be heteroscedastic. We split the data into two subsets at time k and estimate the autoregression parameter from both subsamples. If the difference between the estimates is large for some k, we reject the stability of the data. We discuss the effect of heteroscedasticity on the limit processes.

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

Detecting Changes in the Regression Parameter of AR(1) and RCA(1) Processes with Changing Errors

  • 章节

  • 发表日期：

    2026年

查看全文
  • Lajos Horváth

摘要

We investigate change points in the autoregression parameters of AR(1) and RCA(1) sequences. The errors are allowed to be heteroscedastic. We split the data into two subsets at time k and estimate the autoregression parameter from both subsamples. If the difference between the estimates is large for some k, we reject the stability of the data. We discuss the effect of heteroscedasticity on the limit processes.