A new family of robust regression estimators utilizing robust regression tools and supplementary attributes

Sajjad I., Hanif M., KOYUNCU N., Shahzad U., Al-Noor N. H.

Statistics in Transition New Series, vol.22, no.1, pp.207-216, 2021 (Scopus) identifier

  • Publication Type: Article / Article
  • Volume: 22 Issue: 1
  • Publication Date: 2021
  • Doi Number: 10.21307/stattrans-2021-012
  • Journal Name: Statistics in Transition New Series
  • Journal Indexes: Scopus, International Bibliography of Social Sciences, Central & Eastern European Academic Source (CEEAS), Directory of Open Access Journals
  • Page Numbers: pp.207-216
  • Keywords: Percentage relative efficiency, Ratio-type estimators, Robust regression tools, SRS, Supplementary attributes
  • Hacettepe University Affiliated: Yes

Abstract

Zaman and Bulut (2018a) developed a class of estimators for a population mean utilising LMS robust regression and supplementary attributes. In this paper, a family of estimators is proposed, based on the adaptation of the estimators presented by Zaman (2019), followed by the introduction of a new family of regression-type estimators utilising robust regression tools (LAD, H-M, LMS, H-MM, Hampel-M, Tukey-M, LTS) and supplementary attributes. The mean square error expressions of the adapted and proposed families are determined through a general formula. The study demonstrates that the adapted class of the Zaman (2019) estimators is in every case more proficient than that of Zaman and Bulut (2018a). In addition, the proposed robust regression estimators based on robust regression tools and supplementary attributes are more efficient than those of Zaman and Bulut (2018a) and Zaman (2019).The theoretical findings are supported by real-life examples.