Robust Neutrosophic Ratio-Type Estimators Using REWLSE: A Simulation-Based Approach for Efficient Mean Estimation under Outlier-Contaminated Data

Abstract

The exact estimations of population mean under the influence of indeterminacy and data contamination are a long-standing issue in survey sampling. Traditional ratio-type estimators are highly sensitive to influential observations, and the neutrosophic methods that are currently used do not effectively describe robustness in the face of uncertainty. The current research constructs a generalized family of neutrosophic robust ratio-type estimators that are developed in the context of Robust and Efficient Weighted Least Square Estimation (REWLSE) framework. Bias and mean square error (MSE) expressions are analytically derived for Ordinary Least Squares (OLS) and REWLSE frameworks in order to allow extensive comparisons between theory and efficiency. Monte Carlo simulations on neutrosophic data are systematically used to study the finitesample behavior of proposed estimators, and an empirical evaluation of these estimators is done using actual temperature data. The simulation and empirical evidence have repeatedly shown that suggested REWLSEbased neutrosophic estimators have significant efficiencies, they remain highly resistant to outliers, and perform better than OLS-based ones. These results support the effectiveness of the suggested framework and highlight its potential to become a powerful and trustworthy alternative to population mean estimation in uncertain, imprecise, and contaminated data environments.OPEN ACCESS Received: 13/10/2025 Accepted: 12/11/2025 Published: 23/01/2026

Document