Estimation of entropy using random sampling

摘要

In this paper, three new entropy estimators of continuous random variables are proposed using simple random sampling (SRS), ranked set sampling (RSS) and double ranked set sampling (DRSS) techniques. The new estimators are obtained by modifying the estimators suggested by Noughabi and Arghami (2010) and Ebrahim et al. (1994). In terms of the root mean square error (RMSEs) and bias values, a numerical comparison is considered to compare the suggested estimators with Vasicek’s (1976) estimator. Our results reveal that the suggested estimators have smaller mean squared error than Vasicek’s estimator. Also, the suggested estimators under double ranked set sampling are more efficient than other suggested estimators based on SRS and RSS.