Abstract
This paper suggests a class of estimators for estimating the finite population mean of the study variable using known population mean of the auxiliary variable . Asymptotic expressions of bias and variance of the suggested class of estimators have been obtained. Asymptotic optimum estimator (AOE) in the class is identified along with its variance formula. It has been shown that the proposed class of estimators is more efficient than usual unbiased, usual ratio, usual product, Bahl and Tuteja (1991), and Kadilar and Cingi (2003) estimators under some realistic conditions. An empirical study is carried out to judge the merits of suggested estimator over other competitors practically.
1. Introduction
The literature on survey sampling describes a great variety of techniques for using auxiliary information to obtain more efficient estimators. Ratio, product, and regression methods of estimation are good examples in this context (see Singh [1]). If the correlation between the study variable and the auxiliary variable is positive (high), the ratio method of estimation is quite effective. On the other hand, if this correlation is negative (high) the product method of estimation envisaged by Robson [2] and rediscovered by Murthy [3] can be employed quite effectively. The classical ratio and product estimators are considered to be the most practicable in many situations, but they have the limitation of having at most the same efficiency as that of the linear regression estimator. In the situation where the relation between the study variable and the auxiliary variable is a straight line and passing through the origin, the ratio and product estimators have efficiencies equal to the usual regression estimator. But in many practical situations, the line does not pass through the origin, and in such circumstances the ratio and product estimators do not perform equally well to the regression estimator. Keeping this fact in view, a large number of authors have paid their attention toward the formulation of modified ratio and product estimators, for instance, see Singh [4, 5], Srivastava [6โ8], Reddy [9, 10], Gupta [11], Sahai [12], Sahai and Ray [13], Ray and Sahai [14], Srivenkataramana and Tracy [15, 16], Bandyopadhyay [17], Vos [18], Sisodia and Dwivedi [19], Adhvaryu and Gupta [20], Chaubey et al. [21], Mohanty and Sahoo [22], Singh and Shukla [23], Naik and Gupta [24], Bahl and Tuteja [25], Mohanty and Sahoo [26], Upadhyaya and Singh [27], Singh and Tailor [28], Kadilar and Cingi [29], Singh et al. [30], and others.
In this paper, we have envisaged a new class of estimators for population mean of study variable using information on an auxiliary variable which is highly correlated with the study variable and have shown that the suggested class of estimators is more efficient than some existing estimators.
Consider a finite population of size from which a sample of size is drawn according to simple random sampling without replacement (SRSWOR). Let and be the sample mean estimators of the population means and respectively, of the study variable and the auxiliary variable . Let , , (correlation coefficient between the variables and ) and , where
The remaining part of the paper is organized as follows. Section 2 gives the brief review of some estimators for the population mean of study variable with its properties. In Section 3, a new class of estimators for the population mean is described, and the expressions for the asymptotic bias and variance are obtained. Asymptotic optimum estimator (AOE) in the suggested class is obtained with its variance formula. Section 4 addresses the problem of efficiency comparisons, while in Section 5 an empirical study is carried out to evaluate the performance of different estimators.
2. Reviewing Estimators
It is very well known that the sample mean is an unbiased estimator of population mean , and under (SRSWOR), its variance is given by where .
The usual ratio and product estimators of population mean of the study variable are, respectively, defined as
Bahl and Tuteja [25] suggested exponential ratio-type and product-type estimators for population mean , respectively, as
Kadilar and Cingi [31] suggested a chain ratio-type estimator for population mean as
Following the procedure adopted by Kadilar and Cingi [31], one may define a chain product-type estimator for population mean as To the first degree of approximation, the biases and variances of estimators , , , , , and are, respectively, given by From (2.1) and (2.12)โ(2.17), we have made some efficiency comparisons between the estimators , , , , , , and , as shown in Table 1.
3. Suggested Class of Estimators
We define the following class of estimators for the population mean as where are suitable chosen scalars.
It is to be mentioned that for , the class of estimators reduces to the following class of estimators which is due to Sahai and Ray [13].
While for , it reduces to the new transformed class of estimators defined as
To obtain the bias and variance of suggested class of estimators , we write
such that Expressing (3.1) in terms of โs, we have We assume that , so that , , and are expandable. Now expanding the right-hand side of (3.6), we have Neglecting the terms of โs having greater than two in (3.7), we have
or Taking expectation of both sides of (3.9), we get the bias of the class of estimators to the first degree of approximation as Squaring both sides of (3.9) and neglecting terms โs having power greater than two, we have Taking expectation of both sides of (3.11), we get the variance of to the first degree of approximation as which is minimum when Thus, the resulting minimum variance of is obtained as The minimum variance of equals to the approximate variance of the usual linear regression estimator defined as where , and .
It is interesting to note that if we set and in (3.12), we get the variances of the classes of estimators and , respectively, as If we assume the value of is specified by (say), then the variances of and are, respectively, given by If the value of is specified , then the variances of the estimators and are, respectively, given by To illustrate our general results, we consider a particular case of the proposed class of estimators with its properties. If we set in (3.1), we get an estimator of population mean as Putting in (3.10) and (3.12), we get the bias and variance of , to the first degree of approximation, respectively, as Expression (3.20) clearly indicates that the proposed estimator is better than conventional unbiased estimator if , a condition which is usually met in survey situations.
4. Efficiency Comparisons
From (3.17), we have or equivalently From (3.18), we have or equivalently
From (2.1), (2.12)โ(2.17), and (3.12), we have(i) or equivalently(ii) or equivalently(iii) or equivalently(iv) or equivalently(v) or equivalently(vi) or equivalently(vii) or equivalently From (2.1), (2.12), (2.14), (2.16), and (3.21), we have made some efficiency comparisons between the estimators , , , , and , as shown in Table 2.
It is clearly indicated from Table 2 that the estimator is more efficient than estimators , , , and if
5. Empirical Study
To judge the merits of the suggested estimator over usual unbiased estimator , usual ratio estimator , Bahl and Tuteja [25] exponential ratio-type estimator and Kadilar and Cingi [31] chain ratio-type estimator we have considered three populations. Descriptions of the populations are given below.
Population I (source: Srivastava [7], page 406)
fiber per plant in jute fiber crops, height, , , , .
Population II (source: Gupta and Shabbir [32])
the level of apple production amount (1 unit = 100 tones), the number of apple trees (1 unit = 100 trees), , , , .
Population III (Source: Kadilar and Cingi [33])
the level of apple production amount, the number of apple trees, , , , .
We have computed the percent relative efficiencies (PREs) of different estimators , , ,โโ, and of the population mean with respect to by using the following formula:
and results are summarized in Table 3.
Table 3 exhibits that the proposed estimator is more efficient than usual unbiased estimator , usual ratio estimator , Bahl and Tuteja [25] estimator , and Kadilar and Cingi [31] estimator in the sense of having the largest PRE in all three populations. In populations II and III, the proposed estimator has the largest gain in efficiency over all the estimators, while in population I, there is marginal gain in efficiency as compared to . It is further noted that the condition has been satisfied in population I (), II (), and III (). Thus, we recommend the proposed estimator for its use in practice wherever the condition is satisfied.