Akademik Veri Yönetim Sistemi
MATHEMATICAL AND COMPUTER MODELLING OF DYNAMICAL SYSTEMS, sa.1, 2025 (SCI-Expanded)
In the context of sample surveys, it is only sometimes feasible to obtain complete and accurate information. Non-response represents a significant challenge in this context. As this is a common occurrence in the context of estimation, various approaches are employed to eliminate it. This paper proposes a new class of estimators constructed by combining non-response and unbiased estimator approaches. The simulation study provides a comprehensive evaluation of the performance of various estimators by considering a wide range of scenarios, including different sample sizes, correlation coefficients, non-response rates, and z-values. This extensive simulation framework has explored multiple conditions and variations to ensure a thorough assessment of estimator performance under different settings. The findings show that each member of the proposed family of estimators consistently exhibits a higher PRE value than all other estimators under the scenarios tested.