Here, we have proposed and considered six new estimators of the population ratio (R) of two
population means in SRSWOR scheme, using a variable transformation of the auxiliary
variable, x. The biases and mean squared errors of the proposed estimators were obtained up
to first order approximation. Conditions under which the proposed estimators perform better
than the customary ratio estimator ( x / y Rˆ ) were derived. Also obtained were the
optimality conditions under which some of the proposed estimators could become the best
(optimum) estimators. The results of the study were supported and illustrated numerically.
The empirical illustration confirmed, among other things, both the optimality and efficiency
conditions, which we had earlier obtained theoretically in the study. The empirical study
revealed that relatively large gains in efficiency over the customary ratio estimator could be
obtained by using some of the new estimators proposed in the present study. Again, the
direction (positive or negative) of the linear relationship between the two variables plays a
role in identifying some of the proposed estimators that are likely to be more efficient than
the others, for a given set of data. The first three proposed estimators make use of the sample
mean, x as the lead statistic in the denominator, and are likely to perform better than the last
three proposed estimators, when there is a strong positive linear relationship between the two
variables. When there is a strong negative correlation between the variables, the last three
proposed estimators, which incidentally make use of the transformed sample mean, x* , as the
lead statistic in the denominator are likely to be more efficient than the first three proposed
estimators. However, the best estimators to use for any given set of data could be obtained by