On the Bias and Mean Square Error of the Ratio Estimator

Abstract In deriving the bias and mean square error of the ratio estimator , it is generally assumed that for all possible samples. This assumption is, however, unlikely to be satisfied except in special situations. This article presents a justification for the use of the usual approximations to the bias and mean square error of . Bounds for the deviations of the approximations from the true values are also given.