Robust Regression-Based Ratio-Type Estimators in Simple, Ranked Set, and Median Ranked Set Sampling: Applications in Sustainable Agriculture

Irfan, Muhammad; Ahmad, Farooq; Javed, Maria; Bhatti, Sajjad Haider; Hussain, Muhammad Ali

doi:10.24200/sci.2026.66664.10176

Robust Regression-Based Ratio-Type Estimators in Simple, Ranked Set, and Median Ranked Set Sampling: Applications in Sustainable Agriculture

Articles in Press

Document Type : Research Article

Authors

Muhammad Irfan ¹

Farooq Ahmad ¹

Maria Javed ²

Sajjad Haider Bhatti ³

Muhammad Ali Hussain ⁴

¹ Department of Statistics, Government College University Faisalabad, Pakistan.

² Department of Statistics, Government College University, Faisalabad, Pakistan.

³ College of Statistical Sciences, University of the Punjab, Lahore.

⁴ Business School, NingboTech University, Ningbo, 315100, Zhejiang, China.

10.24200/sci.2026.66664.10176

Abstract

Recently, researchers have developed robust ratio-type estimators for estimating the population mean under simple random sampling. These estimators exploit robust regression techniques such as Least Absolute Deviation (LAD), Huber-M, Huber-MM, Tukey-M, Least Trimmed Squares (LTS), and Least Median of Squares (LMS). The present study aims to introduce a novel and efficient class of robust regression-based ratio-type estimators for estimating the finite population mean under simple random sampling (SRS), ranked set sampling (RSS), and median ranked set sampling (MRSS) schemes. The proposed estimators exhibit superior reliability and efficiency compared to competing methods across all three sampling designs. The mathematical properties of the proposed estimators are analytically investigated, including derivations of bias and mean squared error. Furthermore, Monte Carlo simulation studies are conducted using two real agricultural field populations to assess the empirical performance of the proposed estimators. The numerical results clearly demonstrate the effectiveness and improved efficiency of the proposed estimators relative to existing estimators.

Keywords

Auxiliary information

mean squared error

median ranked set sampling

ranked set sampling

robust regression methods

simple random sampling

Subjects