Neutrosophic Regression Type Estimator for Finite Population Parameters 

Abstract Researches under classical statistics often relies on precise, determinate data to estimate population parameters. However, in certain situations, data may be indeterminate or imprecise. Neutrosophic statistics, a generalization of classical statistics, has been introduced to address these challenges by handling vague, indeterminate, and uncertain information effectively. Several estimators, including ratio estimators, have been proposed in neutrosophic statistics. These ratio estimators perform well when the correlation between the auxiliary and study variables is strong. However, in this study, regression-type estimators were developed, demonstrating superior performance in cases where the correlation between the study and auxiliary variables is high, weak, or moderate. The performance of the proposed estimator was evaluated using simulated data as well as four real-world datasets with indeterminate data, including blood pressure, temperature, natural growth rate, and solar energy data. The proposed neutrosophic regression estimator consistently performed better than existing neutrosophic ratio estimator, modified neutrosophic ratio estimators and Neutrosophic exponential ratio estimator as indicated by performance measures mean square error (MSE) and percent relative efficiency (PRE). This paper highlights the advantages of the neutrosophic regression estimator in improving estimation accuracy when dealing with uncertain and ambiguous data with any range of correlation between the study and the auxiliary variables considered under the study.