Use of auxiliary variables in searching efficient estimator of population mean

The paper deals with the improvement in Yadav et al. (2016) estimator of the population mean using the known coefficient of kurtosis and median of the auxiliary variable. The large sample properties of the estimator, bias and mean squared error (MSE), have been calculated up to the first order of approximation. The optimum values of the characterising scalars which minimise the MSE of the proposed estimator have been obtained. A comparative study has been conducted with the existing estimators of the population mean using auxiliary variable under simple random sampling scheme. To justify the improvement of proposed estimator over Yadav et al. (2016) and other estimators of the population mean, an empirical study is also presented by calculating the mean squared errors of the different estimator of the population mean under simple random sampling.