Allocation Strategies for Estimation of Population Mean in Stratified Random Sampling

This study is an extended version of the work published in Kumar and Vishwakarma (Proceedings of the National Academy of Sciences, India, Section A: Physical Sciences, 90(5): 933-939, 2020). The present study provides theoretical, as well as empirical, investigations of various sample allocation strategies for the estimation of population mean in stratified random sampling. The mathematical expressions for mean square errors (MSEs) of several well-known estimators of population mean are derived under the various allocation schemes considered under investigation. The pre-existing estimators are compared with that of the proposed classes of estimators using the MSE criterion, and the necessary and sufficient conditions (NASCs) for the dominance of proposed classes of estimators are revealed. Furthermore, the empirical results of the study exhibit the superiority of Neyman allocation scheme over the Equal and Proportional allocation schemes for the considered estimators.