Estimator of Population Mean in Stratified Random Sampling : Searls Approach

The study proposes two new population mean estimators using the coefficient of variation in stratified random sampling without replacement. The first estimator without an auxiliary variable is developed from Searls estimator and the second estimator is a ratio estimator developed from Jitthavech and Lorchirachoonkul ratio estimator. The conditions for the efficiency of the first proposed estimator higher than the conventional estimator without auxiliary variable and for the efficiency of the second proposed estimator higher than the conventional separate ratio estimator are also analyzed. The simulation confirms the consistency of the two conditions. Moreover, it can be concluded from the simulation that the efficiency of the second estimator is higher than the separate ratio estimator developed from Sisodia and Dwivedi ratio estimator with the lower biases. Keywords: Searls estimator, Jitthavech and Lorchirachoonkul ratio estimator, stratified random sampling without  replacement, relative efficiency.