Median Estimation Using Non-Conventional Measures of Auxiliary Variable Under Stratified Random Sampling
摘要
This paper presents a comprehensive exploration of the estimation of population median through the use of robust measures within the framework of stratified random sampling. Extending the work of Irfan et al. [11], the proposed estimators aim to improve the accuracy of population median estimation across various strata. We derived explicit expressions for the MSE (Mean Squared Error) specifically within the context of stratified random sampling. The performance of these estimators is rigorously evaluated through the analysis of mean squared error (MSE) expressions, providing a solid mathematical foundation for their efficacy. Further, we extend our study to assess the performance using relative root mean squared error (RRMSE), offering deeper insights into the estimator’s reliability and efficiency. A comprehensive empirical analysis is conducted using real-world datasets from the fields of education, agricultural economics, and marketing to illustrate the practical applicability and effectiveness of the proposed methodology. The results indicate that the suggested estimators perform better than existing ones under certain conditions, offering practical value for median estimation in stratified populations.