On Estimating Finite Population Mean in Simple and Stratified Random Sampling

In this article, we propose an exponential ratio type estimator for estimating the finite population mean in simple and stratified random sampling. The properties of the proposed estimator are obtained and comparison is made with some of the existing estimators. The proposed estimator is found to perform better than the usual mean, ratio, exponential ratio, traditional regression and Pandy (1980 Pandy , G. S. ( 1980 ). Product-cum-power estimators . Cal. Statist. Assoc. Bull. 29 : 103 – 108 . [Google Scholar]) estimators in simple and stratified random sampling. We use six data sets for simple random sampling case and two data sets for stratified random sampling case to compare the performances of all of the estimators considered here.