Ratio estimators for the co-efficient of variation in a finite population

The co-efficient of variation (c.v) is a relative measure of dispersion and is free from unit of measurement. hence it is widely used by the scientists in the desciplines of agriculture,biology,economics and environmental science. in this paper three ratio and three product estimators of c.v are proposed in finite population and small sample performance of the estimators is carried out using bivariate normal distribution. in addition to the small sample mean square error,the performance of the estimators are assessed using the criterion i) mean square error ii) coverage probability and iii) average length of the confidence interval. such type of model based comparison of the estimators of c.v has not been done in the past and thus is more comprehensive in nature. certain guidelines for the use of ratio/product estimators of c.v are provided for the benefit of the applied researchers.