Utilising bivariate auxiliary information for enhanced estimation of population mean under simple and stratified random sampling schemes

The present work suggests some difference-cum-exponential ratio-type estimators to deal with the problem of estimation for population mean. The suggested estimators are based on the linear combination of two auxiliary variables under simple and stratified random sampling schemes. Expressions for the bias, mean squared error (MSE) and minimum MSE of the suggested estimators are derived up to the first degree of approximation. Different real life datasets are used to show the superiorities in terms of percent relative efficiencies (PREs) of the new estimators. The suggested estimators are more efficient as they provide maximum gain in PREs as compared to the traditional and competing estimators under study.