Improved family of estimators of population coefficient of variation under simple random sampling

The coefficient of variation (CV) is a unit less relative measure of dispersion that is commonly employed in scientific and social investigations. In the present article, for elevated estimation of CV, we suggest a novel class of estimators of population CV of the study variable utilizing the known information on an auxiliary variable. Up to the first order of approximation, formulas for the bias and Mean squared Errors (MSE) of the proposed estimators are obtained. The optimum values of the characterizing constants are found, as well as the lowest MSEs corresponding to these values. The efficiencies of proposed and competing estimators are evaluated by comparing their MSEs. Three real and two simulated data sets are used to verify the efficiency conditions. For practical utility in many applications, the estimator with a lower MSE or a greater percentage relative efficiency is preferred. Results show an improvement over the competing estimators.