Scholarlink

<p>The <Emphasis FontCategory="NonProportional">lm()</Emphasis> function in R’s stats package is commonly used for fitting linear regression models, but cannot be used for periodic coefficients regression models proposed by Regui et al. (2024). Similarly, the <Emphasis FontCategory="NonProportional">perARMA</Emphasis> package is not designed to model relationships between independent and dependent variables with periodic coefficients regression models. To address these limitations, we present a new R package, <Emphasis FontCategory="NonProportional">PerRegMod</Emphasis>, designed specifically for fitting periodic coefficients linear regression models. <Emphasis FontCategory="NonProportional">PerRegMod</Emphasis> provides a test to detect periodicity of parameters of regression model and proposes the least squares estimator (LSE) for estimating periodic coefficients linear regression parameters. This paper shows the performance of periodic coefficients regression model compared with <i>S</i>-multiple regression model.</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

PerRegMod: An R Package for Periodic Coefficients Linear Regression Models

  • 期刊论文

  • 发表日期：

    2026年1月19日

查看全文
  • Slimane Regui,
  • Abdelhadi Akharif,
  • Amal Mellouk

摘要

The lm() function in R’s stats package is commonly used for fitting linear regression models, but cannot be used for periodic coefficients regression models proposed by Regui et al. (2024). Similarly, the perARMA package is not designed to model relationships between independent and dependent variables with periodic coefficients regression models. To address these limitations, we present a new R package, PerRegMod, designed specifically for fitting periodic coefficients linear regression models. PerRegMod provides a test to detect periodicity of parameters of regression model and proposes the least squares estimator (LSE) for estimating periodic coefficients linear regression parameters. This paper shows the performance of periodic coefficients regression model compared with S-multiple regression model.