The auxiliary information is frequently used to increase precision of the estimates by taking advantage of correlation between
the study variable and the auxiliary variable. Another way to increase the efficiency of the ratio estimator is to use the
information on the auxiliary attributes. Several authors, including Koyuncu [3], Shabbir and Gupta [6], [7] Naik and Gupta
[5], Abd-Elfattah et al. [1], have proposed improved estimators of finite population mean using information on an auxiliary
attribute.
Consider X ¼ fX1; X2; :::; Xi:::; XNg be a finite population of size N. A sample of size n is selected from X by using simple
random sampling without replacement. Let yi and ui denote the values of the study variable and the binary auxiliary attribute
for the ith unit of the population, respectively. Here it is assumed that ui can take only two possible values, depending
on the presence of an attribute, say u, i.e.,
ui ¼ 1; if the ith unit of the population possesses attribute u;
ui ¼ 0; otherwise:
Let P ¼ PN
i¼1ui and p ¼ Pn
i¼1ui denote the total number of units in the population and in the sample, respectively,
possessing an auxiliary attribute u. The corresponding population and sample proportions are P ¼ P
N and p ¼ p
n, respectively.
Similarly, let Y ¼ 1
N
PN
i¼1yi and y ¼ 1
n
Pn
i¼1yi be the population and the sample means of the study variable y, respectively. In
order to estimate the population mean Y, we assume that P is known. Let s2
y ¼ 1
n1
Pn
i¼1ðyi yÞ
2 and s2
u ¼ npð1pÞ
n1 be the sample
variances corresponding to the population variances S2
y ¼ 1
N1
PN
i¼1ðyi YÞ
2
and S2
u ¼ NPð1PÞ
N1 , respectively. Let qpb be the correlation
coefficient between the study variable y and the auxiliary attribute u. Let Cy ¼ Sy
Y and Cu ¼ Su
P be the coefficients of
variation of y and u, respectively. In order to find the biases and mean squared errors (MSEs) of the estimators, we define the
following relative error terms.
0096-3