This action might not be possible to undo. Are you sure you want to continue?
AUXILIARY INFORMATION
Rajesh Singh
Department of Statistics, BHU, Varanasi (U. P.), India
Editor
Florentin Smarandache
Department of Mathematics, University of New Mexico, Gallup, USA
Editor
Estimator
Bias
MSE
First order
Second order First order Second order
t
1
0.0044915
0.004424
39.217225
39.45222
t
2
0
0.00036
39.217225
(for g=1)
39.33552
(for g=1)
t
3
0.04922
0.04935
39.217225
39.29102
t
4
0.2809243
0.60428
39.217225
39.44855
t
5
0.027679
0.04911
39.217225
39.27187
1
ON IMPROVEMENT IN ESTIMATING POPULATION PARAMETER(S) USING
AUXILIARY INFORMATION
Rajesh Singh
Department of Statistics, BHU, Varanasi (U. P.), India
Editor
Florentin Smarandache
Department of Mathematics, University of New Mexico, Gallup, USA
Editor
Educational Publishing & J ournal of Matter Regularity (Beijing)
2013
2
This book can be ordered on paper or electronic formats from:
Education Publishing
1313 Chesapeake Avenue
Columbus, Ohio 43212
USA
Tel. (614) 4850721
Email: info@edupublisher.com
Copyright 2013 by EducationPublishing & Journal of Matter Regularity (Beijing), Editors
and the Authors for their papers
Front and back covers by the Editors
Peer Reviewers:
Prof. Adrian Olaru, University POLITEHNICA of Bucharest (UPB), Bucharest, Romania.
Dr. Linfan Mao, Academy of Mathematics and Systems, Chinese Academy of Sciences,
Beijing 100190, P. R. China.
Y. P. Liu, Department of Applied Mathematics, Beijing Jiaotong University, Beijing 100044,
P. R. China.
Prof. Xingsen Li, School of Management, Ningbo Institute of Technology, Zhejiang
University, Ningbo 315100, P. R. China.
ISBN: 9781599732305
Printed in the United States of America
3
Contents
Preface: 4
1. Use of auxiliary information for estimating population mean in systematic
sampling under non response: 5
2. Some improved estimators of population mean using information on two
auxiliary attributes: 17
3. Study of some improved ratio type estimators under second order
approximation: 25
4. Improvement in estimating the population mean using dual to ratiocumproduct
estimator in simple random sampling: 42
5. Some improved estimators for population variance using two auxiliary variables
in double sampling: 54
4
Preface
The purpose of writing this book is to suggest some improved estimators using
auxiliary information in sampling schemes like simple random sampling and systematic
sampling.
This volume is a collection of five papers, written by eight coauthors (listed in the
order of the papers): Manoj K. Chaudhary, Sachin Malik, Rajesh Singh, Florentin
Smarandache, Hemant Verma, Prayas Sharma, Olufadi Yunusa, and Viplav Kumar Singh,
from India, Nigeria, and USA.
The following problems have been discussed in the book:
In chapter one an estimator in systematic sampling using auxiliary information is
studied in the presence of nonresponse. In second chapter some improved estimators are
suggested using auxiliary information. In third chapter some improved ratiotype estimators
are suggested and their properties are studied under second order of approximation.
In chapter four and five some estimators are proposed for estimating unknown
population parameter(s) and their properties are studied.
This book will be helpful for the researchers and students who are working in the field
of finite population estimation.
5
Use of Auxiliary Information for Estimating Population Mean in Systematic Sampling
under Non Response
1
Manoj K. Chaudhary,
1
Sachin Malik, †
1
Rajesh Singh and
2
Florentin Smarandache
1
Department of Statistics, Banaras Hindu University, Varanasi221005, India
2
Department of Mathematics, University of New Mexico, Gallup, USA
†Corresponding author
Abstract
In this paper we have adapted Singh and Shukla (1987) estimator in systematic
sampling using auxiliary information in the presence of nonresponse. The properties of the
suggested family have been discussed. Expressions for the bias and mean square error (MSE)
of the suggested family have been derived. The comparative study of the optimum estimator
of the family with ratio, product, dual to ratio and sample mean estimators in systematic
sampling under nonresponse has also been done. One numerical illustration is carried out to
verify the theoretical results.
Keywords: Auxiliary variable, systematic sampling, factortype estimator, mean square
error, nonresponse.
6
1. Introduction
There are some natural populations like forests etc., where it is not possible to apply
easily the simple random sampling or other sampling schemes for estimating the population
characteristics. In such situations, one can easily implement the method of systematic
sampling for selecting a sample from the population. In this sampling scheme, only the first
unit is selected at random, the rest being automatically selected according to a predetermined
pattern. Systematic sampling has been considered in detail by Madow and Madow (1944),
Cochran (1946) and Lahiri (1954). The application of systematic sampling to forest surveys
has been illustrated by Hasel (1942), Finney (1948) and Nair and Bhargava (1951).
The use of auxiliary information has been permeated the important role to improve
the efficiency of the estimators in systematic sampling. Kushwaha and Singh (1989)
suggested a class of almost unbiased ratio and product type estimators for estimating the
population mean using jackknife technique initiated by Quenouille (1956). Later Banarasi et
al. (1993), Singh and Singh (1998), Singh et al. (2012), Singh et al. (2012) and Singh and
Solanki (2012) have made an attempt to improve the estimators of population mean using
auxiliary information in systematic sampling.
The problem of nonresponse is very common in surveys and consequently the
estimators may produce bias results. Hansen and Hurwitz (1946) considered the problem of
estimation of population mean under nonresponse. They proposed a sampling plan that
involves taking a subsample of nonrespondents after the first mail attempt and then
enumerating the subsample by personal interview. ElBadry (1956) extended Hansen and
Hurwitz (1946) technique. Hansen and Hurwitz (1946) technique in simple random sampling
is described as: From a population U = (U1, U2, , UN), a large first phase sample of size n’
is selected by simple random sampling without replacement (SRSWOR). A smaller second
phase sample of size n is selected from n’ by SRSWOR. Nonresponse occurs on the second
phase of size n in which n1 units respond and n2 units do not. From the n2 nonrespondents,
by SRSWOR a sample of r = n2/ k; k > 1units is selected. It is assumed that all the r units
respond this time round. (see Singh and Kumar (20009)). Several authors such as Cochran
(1977), Sodipo and Obisesan (2007), Rao (1987), Khare and Srivastava ( 1997) and Okafor
and Lee (2000) have studied the problem of nonresponse under SRS.
7
In the sequence of improving the estimator, Singh and Shukla (1987) proposed a
family of factortype estimators for estimating the population mean in simple random
sampling using an auxiliary variable, as
( )
( )
(
¸
(
¸
+ +
+ +
=
x C X fB A
x fB X C A
y T
α
(1.1)
where y and x are the sample means of the population means Y and X respectively. A,
B and C are the functions ofα , which is a scalar and chosen so as the MSE of the estimator
Tα is minimum.
Where,
( )( ) 2 1 − − = α α A , ( )( ) 4 1 − − = α α B ,
( )( )( ) 4 3 2 − − − = α α α C ; 0 > α and
N
n
f = .
Remark 1 : If we take α = 1, 2, 3 and 4, the resulting estimators will be ratio, product, dual
to ratio and sample mean estimators of population mean in simple random sampling
respectively (for details see Singh and Shukla (1987) ).
In this paper, we have proposed a family of factortype estimators for estimating the
population mean in systematic sampling in the presence of nonresponse adapting Singh and
Shukla (1987) estimator. The properties of the proposed family have been discussed with the
help of empirical study.
2. Sampling Strategy and Estimation Procedure
Let us assume that a population consists of N units numbered from 1 to N in some
order. If nk N = , where k is a positive integer, then there will be k possible samples each
consisting of n units. We select a sample at random and collect the information from the
units of the selected sample. Let
1
n units in the sample responded and
2
n units did not
respond, so that n n n = +
2 1
. The
1
n units may be regarded as a sample from the response
class and
2
n units as a sample from the nonresponse class belonging to the population. Let
us assume that
1
N and
2
N be the number of units in the response class and nonresponse
8
class respectively in the population. Obviously,
1
N and
2
N are not known but their unbiased
estimates can be obtained from the sample as
n N n N /
ˆ
1 1
= ; n N n N /
ˆ
2 2
= .
Further, using Hansen and Hurwitz (1946) technique we select a subsample of size
2
h from the
2
n nonrespondent units such that L h n
2 2
= ( 1 > L ) and gather the information
on all the units selected in the subsample (for details on Hansen and Hurwitz (1946)
technique see Singh and Kumar (2009)).
Let Y and X be the study and auxiliary variables with respective population means
Y and X . Let ( )
ij ij
x y be the observation on the
th
j unit in the
th
i systematic sample under
study (auxiliary) variable ( n j k i ... 1 : ... 1 = = ).Let us consider the situation in which non
response is observed on study variable and auxiliary variable is free from nonresponse. The
HansenHurwitz (1946) estimator of population mean Y and sample mean estimator of X
based on a systematic sample of size n , are respectively given by
n
y n y n
y
h n
2
2 1 1
* +
=
and
¿
=
=
n
j
ij
x
n
x
1
1
where
1 n
y and
2
h
y are respectively the means based on
1
n respondent units and
2
h non
respondent units. Obviously,
*
y and x are unbiased estimators of Y and X respectively. The
respective variances of
*
y and x are expressed as
( ) ( ) { }
2
2 2
2
* 1
1 1
1
Y Y Y
S W
n
L
S n
nN
N
y V
−
+ − +
−
= ρ (2.1)
and
( ) x V = ( ) { }
2
1 1
1
X X
S n
nN
N
ρ − +
−
(2.2)
9
where
Y
ρ and
X
ρ are the correlation coefficients between a pair of units within the
systematic sample for the study and auxiliary variables respectively.
2
Y
S and
2
X
S are
respectively the mean squares of the entire group for study and auxiliary variables.
2
2 Y
S be the
population mean square of nonresponse group under study variable and
2
W is the non
response rate in the population.
Assuming population mean X of auxiliary variable is known, the usual ratio,
product and dual to ratio estimators based on a systematic sample under nonresponse are
respectively given by
*
R
y = X
x
y
*
, (2.3)
*
P
y =
X
x y
*
(2.4)
and
*
D
y =
( )
( )X n N
x n X N
y
−
− *
. (2.5)
Obviously, all the above estimators
*
R
y ,
*
P
y and
*
D
y are biased. To derive the biases
and mean square errors (MSE) of the estimators
*
R
y ,
*
P
y and
*
D
y under large sample
approximation, let
*
y = ( )
0
1 e Y +
x = ( )
1
1 e X +
such that ( )
0
e E = ( )
1
e E = 0,
( )
2
0
e E =
( )
2
*
Y
y V
= ( ) { }
2
2
2
2
2
1
1 1
1
Y
S
W
n
L
C n
nN
N
Y
Y Y
−
+ − +
−
ρ , (2.6)
( )
2
1
e E =
( )
2
X
x V
= ( ) { }
2
1 1
1
X X
C n
nN
N
ρ − +
−
(2.7)
and
10
( )
1 0
e e E =
( )
X Y
x y Cov ,
*
= ( ) { } ( ) { }
X Y X Y
C C n n
nN
N
ρ ρ ρ
2
1
2
1
1 1 1 1
1
− + − +
−
(2.8)
where
Y
C and
X
C are the coefficients of variation of study and auxiliary variables
respectively in the population (for proof see Singh and Singh(1998) and Singh (2003, pg.
no. 138) ).
The biases and MSE’s of the estimators
*
R
y ,
*
P
y and
*
D
y up to the first order of
approximation using (2.62.8), are respectively given by
( )
*
R
y B = ( ) { }( )
2 *
1 1 1
1
X X
C K n Y
nN
N
ρ ρ − − +
−
, (2.9)
( )
*
R
y MSE = ( ) { } ( ) [ ]
2 * 2 *
2
2 1 1 1
1 2
X Y X
C K C n Y
nN
N
ρ ρ ρ − + − +
−
+
2
2 2
1
Y
S W
n
L −
, (2.10)
( )
*
P
y B = ( ) { }
2 *
1 1
1
X X
C K n Y
nN
N
ρ ρ − +
−
, (2.11)
( )
*
P
y MSE = ( ) { } ( ) [ ]
2 * 2 *
2
2 1 1 1
1 2
X Y X
C K C n Y
nN
N
ρ ρ ρ + + − +
−
+
2
2 2
1
Y
S W
n
L −
, (2.12)
( )
*
D
y B = ( ) { }[ ]
2
X
*
X
C K 1 n 1 Y
nN
1 N
ρ − ρ − +
−
, (2.13)
( )
*
D
y MSE = ( ) { }
(
(
¸
(
¸
)
`
¹
¹
´
¦
−


.

\

−


.

\

−
+ − +
−
2 * 2 *
2
2
1 1
1 1
1 2
X Y X
C K
f
f
f
f
C n Y
nN
N
ρ ρ ρ
+
( )
2
2 2
1
Y
S W
n
L −
(2.14)
where,
*
ρ =
( ) { }
( ) { }
2
1
2
1
1 1
1 1
X
Y
n
n
ρ
ρ
− +
− +
and
X
Y
C
C
K ρ = .
( for details of proof refer to Singh et al.(2012)).
The regression estimator based on a systematic sample under nonresponse is given by
11
) 15 . 2 ( ) x X ( b
*
y y
*
lr
− + =
MSE of the estimator
*
lr
y is given by
) y ( MSE
*
lr
= ( ) { }[ ]
2
* 2 2 2
2
1 1
1
ρ ρ
X Y X
C K C n Y
nN
N
− − +
−
+
( )
2
2 2
1
Y
S W
n
L −
(2.16)
3. Adapted Family of Estimators
Adapting the estimator proposed by Singh and Shukla (1987), a family of factortype
estimators of population mean in systematic sampling under nonresponse is written as
( )
( )
(
¸
(
¸
+ +
+ +
=
x C X fB A
x fB X C A
y T
*
*
α
. (3.1)
The constants A, B, C, and f are same as defined in (1.1).
It can easily be seen that the proposed family generates the nonresponse versions of
some well known estimators of population mean in systematic sampling on putting different
choices ofα . For example, if we take α = 1, 2, 3 and 4, the resulting estimators will be ratio,
product, dual to ratio and sample mean estimators of population mean in systematic sampling
under nonresponse respectively.
3.1 Properties of
*
α
T
Obviously, the proposed family is biased for the population meanY . In order to find
the bias and MSE of
*
α
T , we use large sample approximations. Expressing the equation (3.1)
in terms of
i
e ’s ( ) 1 , 0 = i we have
( )( ) ( ) ( ) [ ]
C fB A
e fB C A De e Y
T
+ +
+ + + + +
=
−
1
1
1 0 *
1 1 1
α
(3.2)
where .
C fB A
C
D
+ +
=
Since 1 < D and 1 <
i
e , neglecting the terms of
i
e ’s ( ) 1 , 0 = i having power greater
than two, the equation (3.2) can be written as
12
Y T −
*
α
= ( ){ } [
1 0
2
1
2
1 0
e De e D De e C A
C fB A
Y
− + − +
+ +
( ) ( ) ( ) { }]
1 0
2
1 1 0
1 1 1 e e D e D D e D e fB − − − + − − + . (3.3)
Taking expectation of both sides of the equation (3.3), we get
[ ] Y T E −
*
α
=
( )
( ) ( )
(
¸
(
¸
−
+ + + +
−
1 0
2
1
e e E e E
C fB A
C
C fB A
fB C Y
.
Let ( ) α φ
1
=
C fB A
fB
+ +
and ( ) α φ
2
=
C fB A
C
+ +
then
( ) α φ = ( ) α φ
2
 ( ) α φ
1
=
C fB A
fB C
+ +
−
.
Thus, we have
[ ] Y T E −
*
α
= ( ) ( ) ( ) ( ) [ ]
1 0
2
1 2
e e E e E Y − α φ α φ . (3.4)
Putting the values of ( )
2
1
e E and ( )
1 0
e e E from equations (2.7) and (2.8) into the
equation (3.4), we get the bias of
*
α
T as
( )
*
α
T B = ( ) ( ) { } ( ) [ ]
2 *
2
1 1
1
X X
C K n Y
nN
N
ρ α φ ρ α φ − − +
−
. (3.5)
Squaring both the sides of the equation (3.3) and then taking expectation, we get
[ ]
2
*
Y T E −
α
= ( ) ( ) ( ) ( ) ( ) [ ]
1 0
2
1
2 2
0
2
2 e e E e E e E Y α φ α φ − + . (3.6)
Substituting the values of ( )
2
0
e E , ( )
2
1
e E and ( )
1 0
e e E from the respective equations
(2.6), (2.7) and (2.8) into the equation (3.6), we get the MSE of
*
α
T as
( )
*
α
T MSE = ( ) { } ( ) ( ) { } [ ]
2 * 2 2 *
2
2 1 1
1 2
X Y X
C K C n Y
nN
N
ρ α φ α φ ρ ρ − + − +
−
+
( )
2
2 2
1
Y
S W
n
L −
. (3.7)
13
3.2 Optimum Choice of α
In order to obtain the optimum choice of α , we differentiate the equation (3.7) with
respect to α and equating the derivative to zero, we get the normal equation as
( ) { } ( ) ( ) ( ) [ ]
2 *
2
2 2 1 1
1
X X
C K n Y
nN
N
ρ α φ α φ α φ ρ ′ − ′ − +
−
= 0 (3.8)
where ( ) α φ′ is the first derivative of ( ) α φ with respect to α .
Now from equation (3.8), we get
( ) α φ = K
*
ρ (3.9)
which is the cubic equation inα . Thus α has three real roots for which the MSE of proposed
family would attain its minimum.
Putting the value of ( ) α φ from equation (3.9) into equation (3.7), we get
( )
min
*
α
T MSE = ( ) { }[ ]
2
* 2 2 2
2
1 1
1
ρ ρ
X Y X
C K C n Y
nN
N
− − +
−
+
( )
2
2 2
1
Y
S W
n
L −
(3.10)
which is the MSE of the usual regression estimator of population mean in systematic
sampling under nonresponse.
4. Empirical Study
In the support of theoretical results, we have considered the data given in Murthy
(1967, p. 131132). These data are related to the length and timber volume for ten blocks of
the blacks mountain experimental forest. The value of intraclass correlation coefficients
X
ρ and
Y
ρ have been given approximately equal by Murthy (1967, p. 149) and Kushwaha
and Singh (1989) for the systematic sample of size 16 by enumerating all possible systematic
samples after arranging the data in ascending order of strip length. The particulars of the
population are given below:
N = 176, n = 16, Y = 282.6136, X = 6.9943,
2
Y
S = 24114.6700,
2
X
S = 8.7600, ρ = 0.8710,
14
2
2 Y
S =
4
3
2
Y
S = 18086.0025.
Table 1 depicts the MSE’s and variance of the estimators of proposed family with
respect to nonresponse rate (
2
W ).
Table 1: MSE and Variance of the Estimators for L = 2.
α
2
W
0.1 0.2 0.3 0.4
1 (=
*
R
y )
371.37 484.41 597.45 710.48
2 (=
*
P
y )
1908.81 2021.85 2134.89 2247.93
3(=
*
D
y )
1063.22 1176.26 1289.30 1402.33
4(=
*
y )
1140.69 1253.13 1366.17 1479.205
) ) ( (
min
*
α
α T
opt
=
270.67 383.71 496.75 609.78
5. Conclusion
In this paper, we have adapted Singh and Shukla (1987) estimator in systematic
sampling in the presence of nonresponse using an auxiliary variable and obtained the
optimum estimator of the proposed family. It is observed that the proposed family can
generate the nonresponse versions of a number of estimators of population mean in
systematic sampling on different choice ofα . From Table 1, we observe that the proposed
family under optimum condition has minimum MSE, which is equal to the MSE of the
regression estimator (most of the class of estimators in sampling literature under optimum
condition attains MSE equal to the MSE of the regression estimator). It is also seen that the
MSE or variance of the estimators increases with increase in non response rate in the
population.
15
References
1. Banarasi, Kushwaha, S.N.S. and Kushwaha, K.S. (1993): A class of ratio, product and
difference (RPD) estimators in systematic sampling, Microelectron. Reliab., 33, 4,
455–457.
2. Cochran, W. G. (1946): Relative accuracy of systematic and stratified random
samples for a certain class of population, AMS, 17, 164177.
3. Finney, D.J. (1948): Random and systematic sampling in timber surveys, Forestry,
22, 6499.
4. Hansen, M. H. and Hurwitz, W. N. (1946) : The problem of nonresponse in sample
surveys, Jour. of The Amer. Stat. Assoc., 41, 517529.
5. Hasel, A. A. (1942): Estimation of volume in timber stands by strip sampling, AMS,
13, 179206.
6. Kushwaha, K. S. and Singh, H.P. (1989): Class of almost unbiased ratio and product
estimators in systematic sampling, Jour. Ind. Soc. Ag. Statistics, 41, 2, 193–205.
7. Lahiri, D. B. (1954): On the question of bias of systematic sampling, Proceedings of
World Population Conference, 6, 349362.
8. Madow, W. G. and Madow, L.H. (1944): On the theory of systematic sampling, I.
Ann. Math. Statist., 15, 124.
9. Murthy, M.N. (1967): Sampling Theory and Methods. Statistical Publishing Society,
Calcutta.
10. Nair, K. R. and Bhargava, R. P. (1951): Statistical sampling in timber surveys in
India, Forest Research Institute, Dehradun, Indian forest leaflet, 153.
11. Quenouille, M. H. (1956): Notes on bias in estimation, Biometrika, 43, 353360.
12. Singh, R and Singh, H. P. (1998): Almost unbiased ratio and product type estimators
in systematic sampling, Questiio, 22,3, 403416.
13. Singh, R., Malik, S., Chaudhary, M.K., Verma, H. and Adewara, A. A. (2012) : A
general family of ratio type estimators in systematic sampling. Jour. Reliab. and Stat.
Stud.,5(1), 7382).
14. Singh, R., Malik, S., Singh, V. K. (2012) : An improved estimator in systematic
sampling. Jour. Of Scie. Res., 56, 177182.
16
15. Singh, H.P. and Kumar, S. (2009) : A general class of dss estimators of population
ratio, product and mean in the presence of nonresponse based on the subsampling of
the nonrespondents. Pak J. Statist., 26(1), 203238.
16. Singh, H.P. and Solanki, R. S. (2012) : An efficient class of estimators for the
population mean using auxiliary information in systematic sampling. Jour. of Stat.
Ther. and Pract., 6(2), 274285.
17. Singh, S. (2003) : Advanced sampling theory with applications. Kluwer Academic
Publishers.
18. Singh, V. K. and Shukla, D. (1987): One parameter family of factortype ratio
estimators, Metron, 45 (12), 273283.
17
Some Improved Estimators of Population Mean Using Information on Two
Auxiliary Attributes
1
Hemant Verma, †
1
Rajesh Singh and
2
Florentin Smarandache
1
Department of Statistics, Banaras Hindu University, Varanasi221005, India
2
Department of Mathematics, University of New Mexico, Gallup, USA
†Corresponding author
Abstract
In this paper, we have studied the problem of estimating the finite population mean
when information on two auxiliary attributes are available. Some improved estimators in
simple random sampling without replacement have been suggested and their properties are
studied. The expressions of mean squared error’s (MSE’s) up to the first order of
approximation are derived. An empirical study is carried out to judge the best estimator out of
the suggested estimators.
Key words: Simple random sampling, auxiliary attribute, point biserial correlation, phi
correlation, efficiency.
Introduction
The role of auxiliary information in survey sampling is to increase the precision of
estimators when study variable is highly correlated with auxiliary variable. But when we talk
about qualitative phenomena of any object then we use auxiliary attributes instead of
auxiliary variable. For example, if we talk about height of a person then sex will be a good
auxiliary attribute and similarly if we talk about particular breed of cow then in this case milk
produced by them will be good auxiliary variable.
Most of the times, we see that instead of one auxiliary variable we have
information on two auxiliary variables e.g.; to estimate the hourly wages we can use the
information on marital status and region of residence (see Gujrati and Sangeetha (2007),
page311).
18
In this paper, we assume that both auxiliary attributes have significant point biserial
correlation with the study variable and there is significant phicorrelation (see Yule (1912))
between the auxiliary attributes.
Consider a sample of size n drawn by simple random sampling without replacement
(SRSWOR) from a population of size N. let yj,
ij
φ (i=1,2) denote the observations on variable
y and
i
φ (i=1,2) respectively for the j
th
unit (i=1,2,3,……N) . We note that
ij
φ =1, if j
th
unit
possesses attribute
ij
φ =0 otherwise . Let , A
N
1 j
ij i ¿
=
φ =
¿
=
φ =
n
1 j
ij i
a ; i=1,2 denotes the total
number of units in the population and sample respectively, possessing attributeφ . Similarly,
let
N
A
P
i
i
= and
n
a
p
i
i
= ;(i=1,2 ) denotes the proportion of units in the population and
sample respectively possessing attribute
i
φ (i=1,2).
In order to have an estimate of the study variable y, assuming the knowledge of
the population proportion P, Naik and Gupta (1996) and Singh et al. (2007) respectively
proposed following estimators:


.

\

=
1
1
1
p
P
y t
(1.1)


.

\

=
2
2
2
P
p
y t
(1.2)


.

\

+
−
=
1 1
1
1
3
p P
p P
exp y t
(1.3)


.

\

+
−
=
2 2
2 2
4
P p
P p
exp y t
(1.4)
The bias and MSE expression’s of the estimator’s
i
t (i=1, 2, 3, 4) up to the first order of
approximation are, respectively, given by
( ) [ ]
1 1
pb
2
p 1 1
K 1 C f Y t B − =
(1.5)
( )
C
2
p
pb 1 2
2
2
K f Y t B =
(1.6)
19
( )
(
¸
(
¸
− =
1
1
pb
2
p
1 3
K
4
1
2
C
f Y t B
(1.7)
( )
(
¸
(
¸
+ =
2
2
pb
2
p
1 4
K
4
1
2
C
f Y t B
(1.8)
MSE( ) ( ) [ ]
1 1
pb
2
p
2
y 1
2
1
K 2 1 C C f Y t − + =
(1.9)
MSE( ) ( ) [ ]
2 1
pb
2
p
2
y 1
2
2
K 2 1 C C f Y t + + =
(1.10)
MSE( )
(
¸
(
¸

.

\

− + =
1 1
pb
2
p
2
y 1
2
3
K
4
1
C C f Y t
(1.11)
MSE( )
(
¸
(
¸

.

\

+ + =
2 2
pb
2
p
2
y 1
2
4
K
4
1
C C f Y t
(1.12)
where, ( ) ( )( ), P Y y
1 N
1
S , P
1 N
1
S ,
N
1

n
1
f
N
1 i
j ji i y
2
N
1 i
j ji
2
1
j j
¿ ¿
=
φ
=
φ
− φ −
−
= − φ
−
= =
.
C
C
K ,
C
C
K ), 2 , 1 j ( ;
P
S
C ,
Y
S
C ,
S S
S
2
2 2
1
1 1
j
j
j
j
p
y
pb pb
p
y
pb pb
j
j
p
y
y
y
y
pb
ρ = ρ = = = = = ρ
φ
φ
φ
( )( )
2 1
2 1
2 1
s s
s
and p p
1 n
1
s
n
1 i
2 i 2 1 i 1
φ φ
φ φ
φ
=
φ φ
= ρ − φ − φ
−
=
¿
be the sample phicovariance and phi
correlation between
1
φ and
2
φ respectively, corresponding to the population phicovariance
and phicorrelation ( )( )
¿
=
φ φ
− φ − φ
−
=
N
1 i
2 i 2 1 i 1
P P
1 N
1
S
2 1
.
S S
S
and
2 1
2 1
φ φ
φ φ
φ
ρ =
In this paper we have proposed some improved estimators of population mean using
information on two auxiliary attributes in simple random sampling without replacement. A
comparative study is also carried out to compare the optimum estimators with respect to usual
mean estimator with the help of numerical data.
20
2. Proposed Estimators
Following Olkin (1958), we propose an estimator
1
t as
1. w w such that constants, are w and w where
p
P
w
p
P
w y t
2 1 2 1
2
2
2
1
1
1 5
= +
(
¸
(
¸
+ =
(2.1)
Consider another estimator t6 as
( ) [ ]
constants. are K and K where
p P
p P
exp p P K y K t
62 61
2 2
2 2
1 1 62 61 6 (
¸
(
¸
+
−
− + =
(2.2)
Following Shaoo et al. (1993), we propose another estimator t7 as
( ) ( )
constants. are K and K where
p P K p P K y t
72 71
2 2 72 1 1 71 7
− + − + =
(2.3)
Bias and MSE of estimators
7 6 5
t and t , t :
To obtain the bias and MSE expressions of the estimators ) 7 , 6 , 5 i ( t
i
= to the first
degree of approximation, we define
2
2 2
2
1
1 1
1 0
P
P p
e ,
P
P p
e ,
Y
Y y
e
−
=
−
=
−
=
such that, . 2 , 1 , 0 i ; 0 ) e ( E
i
= =
Also,
, C f ) e ( E , C f ) e ( E , C f ) e ( E
2
p
1
2
2
2
p
1
2
1
2
y 1
2
0
2
1
= = =
, C K f ) e e ( E , C K f ) e e ( E
2
p
pb 1 2 0
2
p
pb 1 1 0
2
2
1
1
= =
, C K f ) e e ( E
2
p
1 2 1
2
φ
=
2
p
p
p
y
pb pb
p
y
pb pb
C
C
K ,
C
C
K ,
C
C
K
1
2
2 2
1
1 1
φ φ
ρ = ρ = ρ =
Expressing (2.1) in terms of e’s we have,
( )
( ) ( )
(
¸
(
¸
+
+
+
+ =
2 2
2
2
1 1
1
1 0 5
e 1 P
P
w
e 1 P
P
w e 1 Y t
21
( ) ( ) ( ) [ ]
1
2 2
1
1 1 0 5
e 1 w e 1 w e 1 Y t
− −
+ + + + =
(3.1)
Expanding the right hand side of (3.1) and retaining terms up to second degrees of e’s, we have,
[ ]
as ion approximat of order first the upto t estimator
of bias get the we sides, both from Y g subtractin then and (3.1) of sides both of ns expectatio Taking
(3.2) e e w e e w e w e w e w e w e 1 Y t
5
2 0 2 1 0 1
2
2 2
2
1 1 2 2 1 1 0 5
− − + + − − + =
( ) ( ) [ ]
2 2 1 1
pb
2
p 2 pb
2
p 1 1 5
K 1 C w K 1 C w f Y ) t ( Bias − + − = (3.3)
From (3.2), we have,
( ) [ ]
2 2 1 1 0 5
e w e w e Y Y t − − ≅ −
(3.4)
Squaring both sides of (3.4) and then taking expectations, we get the MSE of t5 up to the first order of
approximation as
[ ]
2
p 2 1
2
pb pb 2
2
p pb 1
2
p
2
2
2
p
2
1
2
y
1
2
5
2 2 2 1 1 2 1
C K w w 2 C K w 2 C K w 2 C w C w C f Y ) t ( MSE
φ
+ − − + + = (3.5)
Minimization of (3.5) with respect to w1 and w2, we get the optimum values of w1 and w2 , as
( )
[ ]
( ) say w
C K C
K 1 C
C K C
C K C K
 1
w 1 w
say w
C K C
C K C K
w
*
2 2
p
2
p
pb
2
p
2
p
2
p
2
p
2
p pb
) opt ( 1 ) opt ( 2
*
1 2
p
2
p
2
p
2
p pb
) opt ( 1
2 1
1 1
2 1
2 1 1
2 1
2 1 1
=
−
−
=
−
−
=
− =
=
−
−
=
φ
φ
φ
φ
φ
Similarly, we get the bias and MSE expressions of estimator t6 and t7 respectively, as
2
p 1 1 22 pb
2
p 1 61 6
2 2 2
C K f P K
2
1
K
2
1
8
3
C f 1 Y K ) t ( Bias
φ
+
(
¸
(
¸

.

\

− + = (3.6)
0 ) t ( Bias
7
= (3.7)
And
22
( ) ( )

.

\

− =
=

.

\


.

\

− + + =
− + − + =
2
p
2
p pb 1 3
2
p 1 2
pb
2
p
2
y 1 1
2
61 3 1 62 61 2
2
1
2
62 1
2
2
61 6
C K
2
1
C k f A
C f A
K
4
1
C C f 1 A where
3.8 Y K 2 1 A Y P K K 2 A P K A Y K ) t ( MSE
1 1
1
2 2
φ
And the optimum values of
61
K and
62
K are respectively, given as
( )
( )
( ) say K
A A A P
A Y
K
say K
A A A
A
K
*
62 2
3 2 1 1
3
) opt ( 62
*
61 2
3 2 1
2
) opt ( 61
=
−
=
=
−
=
) 9 . 3 ( C K f P P K K 2
C K f Y P K 2 C K f Y P K 2 C f P K C f P K C f Y ) t ( MSE
2
p 1 2 1 72 71
2
p pb 1 2 72
2
p pb 1 1 71
2
p 1
2
2
2
72
2
p 1
2
1
2
71
2
y 1
2
7
2
2 2 1 1 2 1
φ
+
− − + + =
And the optimum values of
72 71
K and K are respectively, given as
( )
( ) say K
C K C
C K K C K
P
Y
K
say K
C K C
C K K C K
P
Y
K
*
72 2
p
2 2
p
2
p pb
2
p pb
2
) opt ( 72
*
71 2
p
2 2
p
2
p pb
2
p pb
1
) opt ( 71
2 1
1 1 1 2
2 1
2 2 1 1
=


.

\

−
−
=
=


.

\

−
−
=
φ
φ
φ
φ
4. Empirical Study
Data: (Source: Government of Pakistan (2004))
The population consists rice cultivation areas in 73 districts of Pakistan. The variables are defined as:
Y= rice production (in 000’ tonnes, with one tonne = 0.984 ton) during 2003,
1
P = production of farms where rice production is more than 20 tonnes during the year 2002, and
2
P = proportion of farms with rice cultivation area more than 20 hectares during the year 2003.
For this data, we have
N=73, Y=61.3,
1
P =0.4247,
2
P =0.3425,
2
y
S =12371.4,
2
1
S
φ
=0.225490,
2
2
S
φ
=0.228311,
1
pb
ρ =0.621,
2
pb
ρ =0.673,
φ
ρ =0.889.
23
The percent relative efficiency (PRE’s) of the estimators ti (i=1,2,…7) with respect to unusual
unbiasedestimator y have been computed and given in Table 4.1.
Table 4.1 : PRE of the estimators with respect to y
Estimator PRE
y
100.00
t1 162.7652
t2 48.7874
t3 131.5899
t4 60.2812
t5 165.8780
t6 197.7008
t7 183.2372
Conclusion
In this paper we have proposed some improved estimators of population mean using
information on two auxiliary attributes in simple random sampling without replacement. From the
Table 4.1 we observe that the estimator t6 is the best followed by the estimator t7 .
References
Government of Pakistan, 2004, Crops Area Production by Districts (Ministry of Food, Agriculture
and Livestock Division, Economic Wing, Pakistan).
Gujarati, D. N. and Sangeetha ( 2007): Basic econometrics. Tata McGraw – Hill.
24
Malik, S. and Singh, R. (2013): A family of estimators of population mean using information on
point biserial and phi correlation coefficient. IJSE
Naik, V. D and Gupta, P.C.(1996): A note on estimation of mean with known population proportion
of an auxiliary character. Jour. Ind. Soc. Agri. Stat., 48(2), 151158.
Olkin, I. (1958): Multivariate ratio estimation for finite populations, Biometrika, 45, 154–165.
Singh, R., Chauhan, P., Sawan, N. and Smarandache, F.( 2007): Auxiliary information and a priori
values in construction of improved estimators. Renaissance High press.
Singh, R., Chauhan, P., Sawan, N. and Smarandache, F. (2008): Ratio Estimators in Simple Random
Sampling Using Information on Auxiliary Attribute. Pak. Jour. Stat. Oper. Res. Vol. IV, No.1, pp. 47
53.
Singh, R., Kumar, M. and Smarandache, F. (2010): Ratio estimators in simple random sampling
when study variable is an attribute. WASJ 11(5): 586589.
Yule, G. U. (1912): On the methods of measuring association between two attributes. Jour. of the
Royal Soc. 75, 579642.
25
Study of Some Improved Ratio Type Estimators Under Second Order
Approximation
1
Prayas Sharma, †
1
Rajesh Singh and
2
Florentin Smarandache
1
Department of Statistics, Banaras Hindu University, Varanasi221005, India
2
Department of Mathematics, University of New Mexico, Gallup, USA
†Corresponding author
Abstract
Chakrabarty (1979), Khoshnevisan et al. (2007), Sahai and Ray (1980), Ismail
et al. (2011) and Solanki et al. (2012) proposed estimators for estimating population mean Y .
Up to the first order of approximation and under optimum conditions, the minimum mean
squared error (MSE) of all the above estimators is equal to the MSE of the regression
estimator. In this paper, we have tried to found out the second order biases and mean square
errors of these estimators using information on auxiliary variable based on simple random
sampling. Finally, we have compared the performance of these estimators with some
numerical illustration.
Keywords: Simple Random Sampling, population mean, study variable, auxiliary variable,
exponential ratio type estimator, exponential product estimator, Bias and MSE.
1. Introduction
Let U= (U1 ,U2 , U3, …..,Ui, ….UN ) denotes a finite population of distinct and
identifiable units. For estimating the population mean Y of a study variable Y, let us
consider X be the auxiliary variable that are correlated with study variable Y, taking the
corresponding values of the units. Let a sample of size n be drawn from this population using
simple random sampling without replacement (SRSWOR) and yi , xi (i=1,2,…..n ) are the
values of the study variable and auxiliary variable respectively for the ith unit of the sample.
26
In sampling theory the use of suitable auxiliary information results in considerable
reduction in MSE of the ratio estimators. Many authors suggested estimators using some
known population parameters of an auxiliary variable. Upadhyaya and Singh (1999), Singh
and Tailor (2003), Kadilar and Cingi (2006), Khoshnevisan et al. (2007), Singh et al. (2007),
Singh et al. (2008) and Singh and Kumar (2011) suggested estimators in simple random
sampling. Most of the authors discussed the properties of estimators along with their first
order bias and MSE. Hossain et al. (2006) studied some estimators in second order
approximation. In this study we have studied properties of some estimators under second
order of approximation.
2. Some Estimators in Simple Random Sampling
For estimating the population mean Y of Y, Chakrabarty (1979) proposed ratio
type estimator 
( )
x
X
y y 1 t
1
α + α − = (2.1)
where
¿
=
=
n
1 i
i
y
n
1
y and . x
n
1
x
n
1 i
i ¿
=
=
Khoshnevisan et al. (2007) ratio type estimator is given by
( )
g
2
X 1 x
X
y t
(
¸
(
¸
β − + β
= (2.2)
where β and g are constants.
Sahai and Ray (1980) proposed an estimator t3 as
(
(
¸
(
¸
)
`
¹
¹
´
¦
− =
W
3
X
x
2 y t
(2.3)
Ismail et al. (2011) proposed and estimator t4 for estimating the population mean Y of Y as
27
( )
( )
p
4
x X b x
x X a x
y t
(
¸
(
¸
− +
− +
= (2.4)
where p, a and b are constant.
Also, for estimating the population mean Y of Y, Solanki et al. (2012) proposed an estimator
t5 as
( )
(
(
¸
(
¸
¦
)
¦
`
¹
¦
¹
¦
´
¦
+
− δ

.

\

− =
λ
) X x (
X x
exp
X
x
2 y t
5
(2.5)
where λ and δ are constants, suitably chosen by minimizing mean square error of the
estimator
5
t .
3. Notations used
Let us define,
Y
Y y
0
e
−
= and
X
X x
1
e
−
= , then )
1
e ( E )
0
e ( E = =0.
For obtaining the bias and MSE the following lemmas will be used:
Lemma 3.1
(i)
02
C
1
L
02
C
n
1
1 N
n N
}
2
)
0
e {( E )
0
e ( V =
−
−
= =
(ii)
20
C
1
L
20
C
n
1
1 N
n N
}
2
)
1
e {( E )
1
e ( V =
−
−
= =
(iii)
11
C
1
L
11
C
n
1
1 N
n N
)}
1
e
0
e {( E )
1
e ,
0
e ( COV =
−
−
= =
Lemma 3.2
(iv)
21
C
2
L
21
C
2
n
1
) 2 N (
) n 2 N (
) 1 N (
) n N (
)}
0
e
2
1
e {( E =
−
−
−
−
=
(v)
30
C
2
L
30
C
2
n
1
) 2 N (
) n 2 N (
) 1 N (
) n N (
)}
3
1
e {( E =
−
−
−
−
=
Lemma 3.3
(vi)
11
C
20
C
4
L 3
31
C
3
L )
0
e
3
1
e ( E + =
(vii)
2
20
C
4
L 3
40
C
3
L
30
C
3
n
1
) 3 N )( 2 N )( 1 N (
)
2
n 6 nN 6 N
2
N ))( n N (
)}
4
1
e {( E + =
− − −
+ − + −
=
(viii)
20
C
4
L 3
40
C
3
L )
2
0
e
2
1
e ( E + =
Where
3
n
1
) 3 N )( 2 N )( 1 N (
)
2
n 6 nN 6 N
2
N ))( n N (
3
L
− − −
+ − + −
= ,
3
4
n
1
) 3 N )( 2 N )( 1 N (
) 1 n )( 1 n N ))( n N ( N
L
− − −
− − − −
=
28
and
q
q
i
p
p
Y
) Y  (Y
X
) X  (Xi
= Cpq .
Proof of these lemma’s are straight forward by using SRSWOR (see Sukhatme and
Sukhatme (1970)).
4. First Order Biases and Mean Squared Errors
The expression for the biases of the estimators t1, t2, t3 ,t4 and t5 are respectively given
by
(
¸
(
¸
α − α =
11 1 20 1 1
C L C L
2
1
Y ) t ( Bias (4.1)
( )
(
¸
(
¸
β −
+
=
11 1 20 1 2
C L g C L
2
1 g g
Y ) t ( Bias (4.2)
( )
(
¸
(
¸
−
−
− =
11 1 20 1 3
C wL C L
2
1 w w
Y ) t ( Bias (4.3)
( )( )
(
¸
(
¸
− −
+ − =
20 1 11 1 20 1 4
C L
2
1 p a b D
C DL C bDL Y ) t ( Bias (4.4)
( )
(
¸
(
¸
−
−
− =
11 1 20 1 5
C KL C L
2
1 K K
Y ) t ( Bias (4.5)
where, D =p (ba) and
( )
2
2
k
λ + δ
= .
Expression for the MSE’s of the estimators t1, t2 t3 t4 and t5 are, respectively given by
[ ]
11 1 20 1
2
02 1
2
1
C L 2 C L C L Y ) t ( MSE α − α + = (4.6)
[ ]
11 1 20 1
2 2
02 1
2
2
C L g 2 C L g C L Y ) t ( MSE β − β + = (4.7)
[ ]
11 1 20 1
2
02 1
2
3
C wL 2 C L w C L Y ) t ( MSE − + = (4.8)
[ ]
11 1 20 1 02 1
2
4
C DL 2 C DL C L Y ) t ( MSE − + = (4.9)
[ ]
11 1 20 1
2
02 1
2
5
C kL 2 C L k C L Y ) t ( MSE − + = (4.10)
5. Second Order Biases and Mean Squared Errors
Expressing estimator ti’s (i=1,2,3,4) in terms of e’s (i=0,1), we get
( ) ( ) ( ) { }
1
1 0 1
e 1 1 e 1 Y t
−
+ α + α − + =
Or
29
)
`
¹
¹
´
¦ α
+
α
−
α
− α + α −
α
+ + = −
4 3
1 0
3 2
1 0 1 0
2
1
1
0 1
e
24
e e
6
e
6
e e e e e
2 2
e
e Y Y t (5.1)
Taking expectations, we get the bias of the estimator
1
t up to the second order of
approximation as
( )
¸
+
α
− α +
α
− α −
α
=
11 20 4 31 3 21 2 30 2 11 1 20 1 1 2
C C L 3 C L
6
C L C L
6
C L C L
2
Y ) t ( Bias
(
¸
(
+
α
+ ) C L 3 C L (
24
2
20 4 40 3
(5.2)
Similarly, we get the biases of the estimator’s t2, t3, t4 and t5 up to second order of
approximation, respectively as
30 2
3
21 2
2
11 1 20 1
2
2 2
C L
6
) 2 g )( 1 g ( g
C L
2
) 1 g ( g
C L g C L
2
) 1 g ( g
Y ) t ( Bias β
+ +
−
¸
β
+
− β − β
+
=
( )
11 20 4 31 3
3
C C L 3 C L
6
) 2 g )( 1 g ( g
+ β
+ +
−
(
¸
(
+ β
+ + +
+ ) C L 3 C L (
24
) 3 g )( 2 g )( 1 g ( g
2
20 4 40 3
4
(5.3)
30 2 21 2 11 1 20 1 3 2
C L
6
) 2 w )( 1 w ( w
C L
2
) 1 w ( w
C wL C L
2
) 1 w ( w
Y ) t ( Bias
− −
−
¸
−
− −
−
− =
( )
11 20 4 31 3
C C L 3 C L
6
) 2 w )( 1 w ( w
+
− −
−
(
¸
(
+
− − −
− ) C L 3 C L (
24
) 3 w )( 2 w )( 1 w ( w
2
20 4 40 3
(5.4)
30 2
2 1
2
21 2
1
11 1 20 1
1
4 2
C L
) D bD 2 D b (
C L
) D bD (
C DL C L
2
) bD D (
Y ) t ( Bias
+ +
−
¸
+
+ − =
( )
11 20 4 31 3 1
2
C C L 3 C L ) bD 2 D b ( + + −
(
¸
(
+
+ + +
+ ) C L 3 C L (
) D bD 3 D b 3 D b (
2
20 4 40 3
3 2 1
2 3
(5.5)
30
where,
2
) 1 p )( a b (
D D
1
− −
=
3
) 2 p )( a b (
D D
1 2
− −
= .
] ) C L 3 C L ( N
2
20 4 40 3
+ − (5.6)
Where, M=
( ) ( )
)
`
¹
¹
´
¦ − λ − λ λ
+ δ
− λ λ
+
δ − δ α
+
δ − δ
3
) 2 )( 1 (
2
) 1 (
4
2 (
24
6
2
1
2 2 3
,
( )
,
2
2
k
λ + δ
=
N=
( ) ( )
.
3
) 3 )( 2 )( 1 (
) 2 (
2
) 1 (
6
6 (
48
12 12
8
1
2
3 2 3 4
)
`
¹
¹
´
¦ − λ − λ − λ λ
+ δ − δ
− λ λ
+
δ − δ α
+
δ + δ − δ
The MSE’s of the estimators t1, t2, t3, t4 and t5 up to the second order of approximation are,
respectively given by
[
21 2
2
30 2
2
11 1 20 1
2
02 1
2
1 2
C L ) 2 ( C L C L 2 C L C L Y ) t ( MSE α + α + α − α − α + =
( )
11 20 4 31 3
2
C C L 3 C L 2 + α −
( )
(
¸
(
+ α + + + + α α + ) C L 3 C L (
24
5
) C C C ( L 3 C L ) 1 (
2
20 4 40 3
2 2
11 02 20 4 22 3
(5.7)
[
21 2
2
30 2
2 3
11 1 20 1
2 2
02 1
2
2 2
C L ) 1 g 3 ( g C L ) 1 g ( g C gL 2 C L g C L Y ) t ( MSE β + + + β − β − β + =
( )
11 20 4 31 3
3
2 3
12 2
C C L 3 C L
3
g 2 g 9 g 7
C gL 2 + β
)
`
¹
¹
´
¦ + +
− β −
( ) ) C C C ( L 3 C L ) 1 g 2 ( g
2
11 02 20 4 22 3
2
+ + β + +
(
¸
(
+ β
)
`
¹
¹
´
¦ + + +
+
2
20 4 40 3
4
2 3
C L 3 C L (
6
3 g 10 g 9 g 2
(5.8)
( )
¸
+ − −
−
− −
−
− =
11 20 4 31 3 30 2 21 2 11 1 20 1 5 2
C C L 3 C L M C ML C L
2
) 1 k ( k
C kL C L
2
) 1 k ( k
Y ) t ( Bias
31
[
12 2 21 2 30 2
2
11 1 20 1
2
02 1
2
3 2
C wL 2 C L ) 1 w ( w C L ) 1 w ( w C wL 2 C L w C L Y ) t ( MSE − + + − − − + =
( )
11 20 4 31 3
2 3
C C L 3 C L
3
w 2 w 3 w 5
+
)
`
¹
¹
´
¦ − −
+
( )
(
¸
(
+
)
`
¹
¹
´
¦ + −
+ + + + ) C L 3 C L (
24
w 11 w 18 w 7
) C C C ( L 3 C L w
2
20 4 40 3
2 3 4
2
11 02 20 4 22 3
(5.9)
[
12 2 21 2
2
1 30 2 1 11 1 20 1
2
02 1
2
4 2
C DL 2 C L ) D 2 D 2 bD 2 ( C L DD 4 C DL 2 C L D C L Y ) t ( MSE − + + + − − + =
{ }( )
11 20 4 31 3
2
1 1
2 2
C C L 3 C L bD 4 bD 4 DD 2 D b 2 D 2 + + + + + +
{ }( ) ) C C C ( L 3 C L bD 2 D 2 D
2
11 02 20 4 22 3 1
2
+ + + + +
{ } ] ) C L 3 C L ( bDD 12 DD 2 D D b 3
2
20 4 40 3 1 2
2
1
2 2
+ + + + + (5.10)
[
30 2
2
12 2 21 2 11 1 20 1
2
02 1
2
5 2
C L ) 1 k ( k C kL 2 C kL C kL 2 C L k C L Y ) t ( MSE − + − + − + =
( ) ( ) ) C C C ( L 3 C L k C C L 3 C L ) 1 k ( k 2
2
11 02 20 4 22 3 11 20 4 31 3
2
+ + + + − +
(
¸
(
+
−
+ ) C L 3 C L (
4
) k k (
2
20 4 40 3
2 2
(5.11)
6. Numerical Illustration
For a natural population data, we have calculated the biases and the mean square error’s
of the estimator’s and compare these biases and MSE’s of the estimator’s under first and
second order of approximations.
Data Set
The data for the empirical analysis are taken from 1981, Utter Pradesh District Census
Handbook, Aligar. The population consist of 340 villages under koil police station, with
Y=Number of agricultural labour in 1981 and X=Area of the villages (in acre) in 1981. The
following values are obtained
32
, 76765 . 73 Y = , 04 . 2419 X = , 120 n , 70 n , 340 N = ′ = = n=70, C02=0.7614, C11=0.2667,
C03=2.6942, C12=0.0747, C12=0.1589, C30=0.7877, C13=0.1321, C31=0.8851, C04=17.4275
C22=0.8424, C40=1.3051
Table 6.1: Biases and MSE’s of the estimators
Estimator
Bias
MSE
First order
Second order First order Second order
t1
0.0044915
0.004424
39.217225
39.45222
t2
0
0.00036
39.217225
(for g=1)
39.33552
(for g=1)
t3
0.04922
0.04935
39.217225
39.29102
t4
0.2809243
0.60428
39.217225
39.44855
t5
0.027679
0.04911
39.217225
39.27187
In the Table 6.1 the biases and MSE’s of the estimators t1, t2, t3, t4 and t5 are written
under first order and second order of approximations. For all the estimators t1, t2, t3, t4 and t5,
it was observed that the value of the biases decreased and the value of the MSE’s increased
for second order approximation. MSE’s of the estimators up to the first order of
approximation under optimum conditions are same. From Table 6.1 we observe that under
33
second order of approximation the estimator t5 is best followed by t3,and t2 among the
estimators considered here for the given data set.
7. Estimators under stratified random sampling
In survey sampling, it is well established that the use of auxiliary information results in
substantial gain in efficiency over the estimators which do not use such information.
However, in planning surveys, the stratified sampling has often proved needful in improving
the precision of estimates over simple random sampling. Assume that the population U
consist of L strata as U=U1, U2,…,UL . Here the size of the stratum Uh is Nh, and the size of
simple random sample in stratum Uh is nh, where h=1, 2,,L.
The Chakrabarty(1979) ratio type estimator under stratified random sampling is given by
( )
st
st st 1
x
X
y y 1 t α + α − = ′
(7.1)
where,
¿
=
=
h
n
1 i
hi
h
h
y
n
1
y , , x
n
1
x
h
n
1 i
hi
h
h ¿
=
=
h
L
1 h
h st
y w y ¿ =
=
,
h
L
1 h
h st
x w x ¿ =
=
, and
. X w X
h
L
1 h
h ¿
=
=
Khoshnevisan et al. (2007) ratio type estimator under stratified random sampling is given by
( )
g
st
st 2
X 1 x
X
y t
(
¸
(
¸
β − + β
= ′
(7.2)
where g is a constant, for g=1 ,
2
t′ is same as conventional ratio estimator whereas for g = 
1, it becomes conventional product type estimator.
Sahai and Ray (1980) estimator t3 under stratified random sampling is given by
(
(
¸
(
¸
)
`
¹
¹
´
¦
− = ′
W
st
st 3
X
x
2 y t
(7.3)
Ismail et al. (2011) estimator under stratified random sampling
4
t′ is given by
34
( )
( )
p
st
st
st 4
x X b x
x X a x
y t
(
¸
(
¸
− +
− +
= ′
(7.4)
Solanki et al. (2012) estimator under stratified random sampling is given as
( )
(
(
¸
(
¸
¦
)
¦
`
¹
¦
¹
¦
´
¦
+
− δ

.

\

− = ′
λ
) X x (
X x
exp
X
x
2 y t
st
st st
st 5
(7.5)
where λ and δ are the constants, suitably chosen by minimizing MSE of the estimator
5
t′ .
8. Notations used under stratified random sampling
Let us define,
y
y y
e
st
0
−
= and
x
x x
e
st
1
−
= , then )
1
e ( E )
0
e ( E = =0.
To obtain the bias and MSE of the proposed estimators, we use the following notations in the
rest of the article:
such that,
and
( ) ( ) [ ]
s
h h
r
h h
L
1 h
s r
h rs
Y y X x E W V − − =
¿
=
+
Also
20
2
L
1 h
2
yh h
2
h
2
0
V
Y
S w
) e ( E =
γ
=
¿
=
02
X
L
1 h
2
xyh h
2
h
2
1
V
S w
) e ( E
2
=
γ
=
¿
=
35
11
Y X
L
1 h
2
xyh h
2
h
1 0
V
S w
) e e ( E =
γ
=
¿
=
Where
1 N
) Y y (
S
h
2
N
1 i
h h
2
yh
h
−
−
=
¿
=
,
1 N
) X x (
S
h
2
N
1 i
h h
2
xh
h
−
−
=
¿
=
,
1 N
) Y y )( X x (
S
h
N
1 i
h h h h
xyh
h
−
−
=
¿
=
−
,
n
f 1
h
h
h
−
= γ ,
N
n
f
h
h
h
= .
n
N
w
h
h
h
=
Some additional notations for second order approximation,
( ) ( ) [ ]
r
h h
s
h h
s r
L
1 h
s r
h rs
X x Y y E
X Y
1
W V − − =
¿
=
+
Where,
( ) ( ) [ ]
¿
=
− − =
h
N
1 i
r
h h
s
h h
h
) h ( rs
X x Y y
N
1
C
¿
=
=
L
1 h
2
) h ( 12 ) h ( 1 3
h 12
X Y
C k
W V
¿
=
=
L
1 h
2
) h ( 21 ) h ( 1 3
h 21
X Y
C k
W V
¿
=
=
L
1 h
3
) h ( 30 ) h ( 1 3
h 30
Y
C k
W V
¿
=
=
L
1 h
3
) h ( 03 ) h ( 1 3
h 03
X
C k
W V
¿
=
+
=
L
1 h
3
) h ( 02 ) h ( 01 ) h ( 3 ) h ( 13 ) h ( 2 4
h 13
X Y
C C k 3 C k
W V
¿
=
+
=
L
1 h
4
2
) h ( 02 ) h ( 3 ) h ( 04 ) h ( 2 4
h 04
X
C k 3 C k
W V
( )
¿
=
+ +
=
L
1 h
2 2
2
) h ( 11 ) h ( 02 ) h ( 01 ) h ( 3 ) h ( 22 ) h ( 2 4
h 22
X Y
C 2 C C k C k
W V
Where
36
) 2 N )( 1 N ( n
) n 2 N )( n N (
k
h h
2
h h h h
) h ( 1
− −
− −
=
) 3 N )( 2 N )( 1 N ( n
) n N ( n 6 N ) 1 N )( n N (
k
h h h
3
h h h h h h h
) h ( 2
− − −
− − + −
=
) 3 N )( 2 N )( 1 N ( n
) 1 n )( 1 n N ( N ) n N (
k
h h h
3
h h h h h h
) h ( 3
− − −
− − − −
=
9. First Order Biases and Mean Squared Errors
The biases of the estimators
1
t′
,
2
t′ ,
3
t′ ,
4
t′
and
5
t′ are respectively given by
(
¸
(
¸
α − α = ′
11 02 1 1
V V L
2
1
Y ) t ( Bias (9.1)
( )
(
¸
(
¸
β − β
+
= ′
11 02
2
2
V g V
2
1 g g
Y ) t ( Bias (9.2)
( )
(
¸
(
¸
−
−
− = ′
11 02 3
wV V
2
1 w w
Y ) t ( Bias (9.3)
( )( )
(
¸
(
¸
− −
+ − = ′
02 11 02 4
V
2
1 p a b D
DV bDV Y ) t ( Bias (9.4)
( )
(
¸
(
¸
−
−
− = ′
11 02 5
KV V
2
1 K K
Y ) t ( Bias (9.5)
Where, D =p(ba) and
( )
2
2
k
λ + δ
= .
The MSE’s of the estimators
1
t′
,
2
t′ ,
3
t′ ,
4
t′
and
5
t′ are respectively given by
[ ]
11 02
2
20
2
1
V 2 V V Y ) t ( MSE α − α + = ′ (9.6)
[ ]
11 02
2 2
20
2
2
V g 2 V g V Y ) t ( MSE β − β + = ′ (9.7)
[ ]
11 02
2
20
2
3
wV 2 V w V Y ) t ( MSE − + = ′ (9.8)
[ ]
11 02 20
2
4
DV 2 DV V Y ) t ( MSE − + = ′ (9.9)
[ ]
11 02
2
20
2
5
kV 2 V k V Y ) t ( MSE − + = ′ (9.10)
37
10. Second Order Biases and Mean Squared Errors
Expressing estimator ti’s (i=1,2,3,4) in terms of e’s (i=0,1), we get
( ) ( ) ( ) { }
1
1 0 1
e 1 1 e 1 Y t
−
+ α + α − + = ′
Or
)
`
¹
¹
´
¦ α
+
α
−
α
− α + α −
α
+ + = − ′
4 3
1 0
3 2
1 0 1 0
2
1
1
0 1
e
24
e e
6
e
6
e e e e e
2 2
e
e Y Y t (10.1)
Taking expectations, we get the bias of the estimator
1
t′ up to the second order of
approximation as
¸
(
¸
( α
+
α
− α +
α
− α −
α
= ′
04 13 12 03 11 02 1 2
V
24
V
6
V V
6
V V
2
Y ) t ( Bias (10.2)
Similarly we get the Biases of the estimator’s
2
t′ ,
3
t′ ,
4
t′
and
5
t′ up to second order of
approximation, respectively as
30
3
12
2
11 02
2
2 2
V
6
) 2 g )( 1 g ( g
V
2
) 1 g ( g
V g V
2
) 1 g ( g
Y ) t ( Bias β
+ +
−
¸
β
+
− β − β
+
= ′
(
¸
(
β
+ + +
+ β
+ +
−
04
4
31
3
V
24
) 3 g )( 2 g )( 1 g ( g
V
6
) 2 g )( 1 g ( g
(10.3)
03 12 11 02 3 2
V
6
) 2 w )( 1 w ( w
V
2
) 1 w ( w
wV V
2
) 1 w ( w
Y ) t ( Bias
− −
−
¸
−
− −
−
− = ′
(
¸
( − − −
−
− −
−
04 31
V
24
) 3 w )( 2 w )( 1 w ( w
V
6
) 2 w )( 1 w ( w
(10.4)
03 2 1
2
12 1 11 02
1
4 2
V ) D bD 2 D b ( V ) D bD ( DV V
2
) bD D (
Y ) t ( Bias + + −
¸
+ + − = ′
]
04 3 2 1
2 3
31 1
2
V ) D bD 3 D b 3 D b ( V ) bD 2 D b ( + + + + + − (10.5)
Where, D =p(ba)
2
) 1 p )( a b (
D D
1
− −
=
3
) 2 p )( a b (
D D
1 2
− −
=
(10.6)
(
¸
(
¸
−
¸
− −
−
− −
−
− = ′
04 31 03 12 11 02 5 2
NV MV MV V
2
) 1 k ( k
kV V
2
) 1 k ( k
Y ) t ( Bias
38
Where, M=
( ) ( )
)
`
¹
¹
´
¦ − λ − λ λ
+ δ
− λ λ
+
δ − δ α
+
δ − δ
3
) 2 )( 1 (
2
) 1 (
4
2 (
24
6
2
1
2 2 3
,
( )
2
2
k
λ + δ
=
N=
( ) ( )
)
`
¹
¹
´
¦ − λ − λ − λ λ
+ δ − δ
− λ λ
+
δ − δ α
+
δ + δ − δ
3
) 3 )( 2 )( 1 (
) 2 (
2
) 1 (
6
6 (
48
12 12
8
1
2
3 2 3 4
Following are the MSE of the estimators
1
t′
,
2
t′ ,
3
t′ ,
4
t′
and
5
t′ up to second order of
approximation
[
12
2
03
2
11 02
2
20
2
1 2
V ) 2 ( V V 2 V V Y ) t ( MSE α + α + α − α − α + = ′
(
¸
(
α + + α α + α −
04
2
22 31
2
V
24
5
V ) 1 ( V 2 (10.7)
[
12
2
03
2 3
11 02
2 2
20
2
2 2
V ) 1 g 3 ( g V ) 1 g ( g gV 2 V g V Y ) t ( MSE β + + + β − β − β + = ′
22
2
31
3
2 3
21
V ) 1 g 2 ( g V
3
g 2 g 9 g 7
gV 2 β + + β
)
`
¹
¹
´
¦ + +
− β −
(
¸
(
β
)
`
¹
¹
´
¦ + + +
+
04
4
2 3
V
6
3 g 10 g 9 g 2
(10.8)
[
21 12 03
2
11 02
2
20
2
3 2
wV 2 V ) 1 w ( w V ) 1 w ( w wV 2 V w V Y ) t ( MSE − + + − − − + = ′
(
¸
(
)
`
¹
¹
´
¦ + −
+ +
)
`
¹
¹
´
¦ − −
+
04
2 3 4
22 31
2 3
V
24
w 11 w 18 w 7
wV V
3
w 2 w 3 w 5
(10.9)
[
21 12
2
1 03 1 11 02
2
20
2
4 2
DV 2 V ) D 2 D 2 bD 2 ( V DD 4 DV 2 V D V Y ) t ( MSE − + + + − − + = ′
{ } { }
22 1
2
31
2
1 1
2 2
V bD 2 D 2 D V bD 4 bD 4 DD 2 D b 2 D 2 + + + + + + + +
{ } ]
04 1 2
2
1
2 2
V bDD 12 DD 2 D D b 3 + + + + (10.10)
[
03
2
21 12 11 02
2
20
2
5 2
V ) 1 k ( k kV 2 kV kV 2 V k V Y ) t ( MSE − + − + − + = ′
(
¸
( −
+ + − +
04
2 2
22 31
2
V
4
) k k (
kV V ) 1 k ( k 2 (10.11)
39
11. Numerical Illustration
For the natural population data, we shall calculate the bias and the mean square error of
the estimator and compare Bias and MSE for the first and second order of approximation.
Data Set1
To illustrate the performance of above estimators, we have considered the natural
Data given in Singh and Chaudhary (1986, p.162).
The data were collected in a pilot survey for estimating the extent of cultivation and
production of fresh fruits in three districts of Uttar Pradesh in the year 19761977.
Table 11.1: Biases and MSE’s of the estimators
Estimator
Bias
MSE
First order Second order First order second order
1
t′
10.82707903
13.65734654
1299.110219
1372.906438
2
t′
10.82707903
6.543275811
1299.110219
1367.548263
3
t′
27.05776113
27.0653128
1299.110219
1417.2785
4
t′
11.69553975
41.84516913
1299.110219
2605.736045
5
t′
22.38574093
14.95110301
1299.110219
2440.644397
40
From Table 11.1 we observe that the MSE’s of the estimators
1
t′
,
2
t′ ,
3
t′ ,
4
t′
and
5
t′ are
same up to the first order of approximation but the biases are different. The MSE of the
estimator
2
t′ is minimum under second order of approximation followed by the estimator
1
t′
and other estimators.
Conclusion
In this study we have considered some estimators whose MSE’s are same up to the
first order of approximation. We have extended the study to second order of approximation
to search for best estimator in the sense of minimum variance. The properties of the
estimators are studied under SRSWOR and stratified random sampling. We have observed
from Table 6.1 and Table 11.1 that the behavior of the estimators changes dramatically
when we consider the terms up to second order of approximation.
REFERENCES
Bahl, S. and Tuteja, R.K. (1991) : Ratio and product type exponential estimator. Information
and Optimization Science XIII 159163.
Chakrabarty, R.P. (1979) : Some ratio estimators, Journal of the Indian Society of
Agricultural Statistics 31(1), 49–57.
Hossain, M.I., Rahman, M.I. and Tareq, M. (2006) : Second order biases and mean squared
errors of some estimators using auxiliary variable. SSRN.
Ismail, M., Shahbaz, M.Q. and Hanif, M. (2011) : A general class of estimator of population
mean in presence of non–response. Pak. J. Statist. 27(4), 467476.
Khoshnevisan, M., Singh, R., Chauhan, P., Sawan, N., and Smarandache, F. (2007). A
general family of estimators for estimating population mean using known value of
some population parameter(s), Far East Journal of Theoretical Statistics 22 181–191.
Ray, S.K. and Sahai, A (1980) : Efficient families of ratio and product type estimators,
Biometrika 67(1) , 211–215.
Singh, D. and Chudhary, F.S. (1986): Theory and analysis of sample survey designs. Wiley
Eastern Limited, New Delhi.
41
Singh, H.P. and Tailor, R. (2003). Use of known correlation coefficient in estimating the
finite population mean. Statistics in Transition 6, 555560.
Singh, R., Cauhan, P., Sawan, N., and Smarandache, F. (2007): Auxiliary Information and A
Priori Values in Construction of Improved Estimators. Renaissance High Press.
Singh, R., Chauhan, P. and Sawan, N. (2008): On linear combination of Ratioproduct type
exponential estimator for estimating finite population mean. Statistics in
Transition,9(1),105115.
Singh, R., Kumar, M. and Smarandache, F. (2008): Almost Unbiased Estimator for
Estimating Population Mean Using Known Value of Some Population Parameter(s).
Pak. J. Stat. Oper. Res., 4(2) pp6376.
Singh, R. and Kumar, M. (2011): A note on transformations on auxiliary variable in survey
sampling. MASA, 6:1, 1719.
Solanki, R.S., Singh, H. P. and Rathour, A. (2012) : An alternative estimator for estimating the
finite population mean using auxiliary information in sample surveys. ISRN
Probability and Statistics doi:10.5402/2012/657682
Srivastava, S.K. (1967) : An estimator using auxiliary information in sample surveys. Cal.
Stat. Ass. Bull. 15:127134.
Sukhatme, P.V. and Sukhatme, B.V. (1970): Sampling theory of surveys with applications.
Iowa State University Press, Ames, U.S.A.
Upadhyaya, L. N. and Singh, H. P. (199): Use of transformed auxiliary variable in estimating
the finite population mean. Biom. Jour., 41, 627636.
42
IMPROVEMENT IN ESTIMATING THE POPULATION MEAN USING DUAL TO
RATIOCUMPRODUCT ESTIMATOR IN SIMPLE RANDOM SAMPLING
1
Olufadi Yunusa,
2
†Rajesh Singh and
3
Florentin Smarandache
1
Department of Statistics and Mathematical Sciences
Kwara State University, P.M.B 1530, Malete, Nigeria
2
Department of Statistics, Banaras Hindu University, Varanasi (U.P.), India
3
Department of Mathematics, University of New Mexico, Gallup, USA
†Corresponding author
ABSTRACT
In this paper, we propose a new estimator for estimating the finite population mean using two
auxiliary variables. The expressions for the bias and mean square error of the suggested
estimator have been obtained to the first degree of approximation and some estimators are
shown to be a particular member of this estimator. Furthermore, comparison of the suggested
estimator with the usual unbiased estimator and other estimators considered in this paper is
carried out. In addition, an empirical study with two natural data from literature is used to
expound the performance of the proposed estimator with respect to others.
Keywords: Dualtoratio estimator; finite population mean; mean square error; multi
auxiliary variable; percent relative efficiency; ratiocumproduct estimator
1. INTRODUCTION
It is well known that the use of auxiliary information in sample survey design results
in efficient estimate of population parameters (e.g. mean) under some realistic conditions.
This information may be used at the design stage (leading, for instance, to stratification,
43
systematic or probability proportional to size sampling designs), at the estimation stage or at
both stages. The literature on survey sampling describes a great variety of techniques for
using auxiliary information by means of ratio, product and regression methods. Ratio and
product type estimators take advantage of the correlation between the auxiliary
variable, x and the study variable, y . For example, when information is available on the
auxiliary variable that is positively (high) correlated with the study variable, the ratio method
of estimation is a suitable estimator to estimate the population mean and when the correlation
is negative the product method of estimation as envisaged by Robson (1957) and Murthy
(1964) is appropriate.
Quite often information on many auxiliary variables is available in the survey which
can be utilized to increase the precision of the estimate. In this situation, Olkin (1958) was the
first author to deal with the problem of estimating the mean of a survey variable when
auxiliary variables are made available. He suggested the use of information on more than one
supplementary characteristic, positively correlated with the study variable, considering a
linear combination of ratio estimators based on each auxiliary variable separately. The
coefficients of the linear combination were determined so as to minimize the variance of the
estimator. Analogously to Olkin, Singh (1967) gave a multivariate expression of Murthy’s
(1964) product estimator, while Raj (1965) suggested a method for using multiauxiliary
variables through a linear combination of single difference estimators. More recently, Abu
Dayyeh et al. (2003), Kadilar and Cingi (2004, 2005), Perri (2004, 2005), Dianna and Perri
(2007), Malik and Singh (2012) among others have suggested estimators for Y using
information on several auxiliary variables.
Motivated by Srivenkataramana (1980), Bandyopadhyay (1980) and Singh et al. (2005)
and with the aim of providing a more efficient estimator; we propose, in this paper, a new
estimator for Y when two auxiliary variables are available.
44
2. BACKGROUND TO THE SUGGESTED ESTIMATOR
Consider a finite population ( )
N
P P P P , ... , ,
2 1
= of N units. Let a sample s of size n
be drawn from this population by simple random sampling without replacements (SRSWOR).
Let
i
y and ) , (
i i
z x represents the value of a response variable y and two auxiliary variables
) , ( z x are available. The units of this finite population are identifiable in the sense that they
are uniquely labeled from 1
to N and the label on each unit is known. Further, suppose in a
survey problem, we are interested in estimating the population mean Y of y , assuming that
the population means ( ) Z X, of ) , ( z x are known. The traditional ratio and product estimators
for Y are given as
x
X
y y
R
= and
Z
z
y y
P
=
respectively, where
¿
=
=
n
i
i
y
n
y
1
1
,
¿
=
=
n
i
i
x
n
x
1
1
and
¿
=
=
n
i
i
z
n
z
1
1
are the sample means of y , x
and z respectively.
Singh (1969) improved the ratio and product method of estimation given above and
suggested the “ratiocumproduct” estimator for Y as
Z
z
x
X
y y
S
=
In literature, it has been shown by various authors; see for example, Reddy (1974) and
Srivenkataramana (1978) that the bias and the mean square error of the ratio estimator
R
y ,
can be reduced with the application of transformation on the auxiliary variable x . Thus,
authors like, Srivenkataramana (1980), Bandyopadhyay (1980) Tracy et al. (1996), Singh et
al. (1998), Singh et al. (2005), Singh et al. (2007), Bartkus and Plikusas (2009) and Singh et
al. (2011) have improved on the ratio, product and ratiocumproduct method of estimation
using the transformation on the auxiliary information. We give below the transformations
employed by these authors:
45
i i
gx X g x − + =
∗
) 1 ( and
i i
gz Z g z − + =
∗
) 1 ( , for N i ..., , 2 , 1 = , (1)
where
n N
n
g
−
= .
Then clearly, x g X g x − + =
∗
) 1 ( and z g Z g z − + =
∗
) 1 ( are also unbiased estimate of
X and Z respectively and ( )
yx
x y Corr ρ − =
∗
, and ( )
yz
z y Corr ρ − =
∗
, . It is to be noted that
by using the transformation above, the construction of the estimators for Y requires the
knowledge of unknown parameters, which restrict the applicability of these estimators. To
overcome this restriction, in practice, information on these parameters can be obtained
approximately from either past experience or pilot sample survey, inexpensively.
The following estimators
∗
R
y ,
∗
P
y and
SE
y are referred to as dual to ratio, product and
ratiocumproduct estimators and are due to Srivenkataramana (1980), Bandyopadhyay
(1980) and Singh et al. (2005) respectively. They are as presented:
X
x
y y
R
∗
∗
= ,
∗
∗
=
z
Z
y y
P
and
∗
∗
=
z
Z
X
x
y y
SE
It is well known that the variance of the simple mean estimator y , under SRSWOR design is
( )
2
y
S y V λ =
and to the first order of approximation, the Mean Square Errors (MSE) of
R
y ,
P
y ,
S
y ,
∗
R
y ,
∗
P
y and
SE
y are, respectively, given by
( ) ( )
yx x y R
S R S R S y MSE
1
2 2
1
2
2 − + = λ
( ) ( )
yz z y P
S R S R S y MSE
2
2 2
2
2
2 + + = λ
( ) [ ] C D S y MSE
y S
+ − = 2
2
λ
( ) ( )
yx x y R
S gR S R g S y MSE
1
2 2
1
2 2
2 − + =
∗
λ
( ) ( )
yz z y P
S gR S R g S y MSE
2
2 2
2
2 2
2 + + =
∗
λ
46
( ) ( ) gD C g S y MSE
y SE
2
2 2
− + = λ
where,
n
f −
=
1
λ ,
N
n
f = , ( )
¿
=
− =
N
i
i y
Y y
N
S
1
2
2
1
, ( )( )
¿
=
− − =
N
i
i i yx
X x Y y
N
S
1
1
,
x y
yx
yx
S S
S
= ρ ,
X
Y
R =
1
,
Z
Y
R =
2
,
2 2
2 2 1
2 2
1
2
z zx x
S R S R R S R C + − = ,
yz yx
S R S R D
2 1
− = and
2
j
S for
) , , ( z y x j = represents the variances of x , y and z respectively; while
yx
S ,
yz
S and
zx
S
denote the covariance between y and x , y and z and z and x respectively. Note that
yz
ρ ,
zx
ρ ,
2
x
S ,
2
z
S ,
yz
S and
zx
S are defined analogously and respective to the subscripts used.
More recently, Sharma and Tailor (2010) proposed a new ratiocumdual to ratio
estimator of finite population mean in simple random sampling, their estimator with its MSE
are respectively given as,
( )
(
¸
(
¸


.

\

− +


.

\

=
∗
X
x
x
X
y y
ST
α α 1
( ) ( )
2 2
1
yx y ST
S y MSE ρ λ − = .
3. PROPOSED DUAL TO RATIOCUMPRODUCT ESTIMATOR
Using the transformation given in (1), we suggest a new estimator for Y as follows:
( )
(
¸
(
¸


.

\

− +


.

\

=
∗
∗ ∗
∗
Z
z
x
X
z
Z
X
x
y y
PR
θ θ 1
We note that when information on the auxiliary variable z is not used (or variable z
takes the value `unity') and 1 = θ , the suggested estimator
PR
y reduces to the `dual to ratio'
estimator
∗
R
y proposed by Srivenkataramana (1980). Also,
PR
y reduces to the `dual to
product' estimator
∗
P
y proposed by Bandyopadhyay (1980) estimator if the information on
the auxiliary variate x is not used and 0 = θ . Furthermore, the suggested estimator reduces
47
to the dual to ratiocumproduct estimator suggested by Singh et al. (2005) when 1 = θ and
information on the two auxiliary variables x and z are been utilized.
In order to study the properties of the suggested estimator
PR
y (e.g. MSE), we write
( )
0
1 k Y y + = ; ( )
1
1 k X x + = ; ( )
2
1 k Z z + = ;
with ( ) ( ) ( ) 0
2 1 0
= = = k E k E k E
and
( )
2
2
2
0
Y
S
k E
y
λ
= ; ( )
2
2
2
1
X
S
k E
x
λ
= ; ( )
2
2
2
2
Z
S
k E
z
λ
= ; ( )
X Y
S
k k E
yx
λ
=
1 0
; ( )
Z Y
S
k k E
yz
λ
=
2 0
;
( )
Z X
S
k k E
zx
λ
=
2 1
,
Now expressing
PR
y in terms of s k' , we have
( ) ( )( ) ( )( ) ( ) [ ]
2
1
1
1
2 1 0
1 1 1 1 1 1 gk gk gk gk k Y y
PR
− − − + − − + =
− −
θ θ (2)
We assume that 1
1
< gk and 1
2
< gk so that the right hand side of (2) is expandable.
Now expanding the right hand side of (2) to the first degree of approximation, we have
( ) ( ) ( ) ( ) [ ]
2
2
2
1 2 1
2
1
2
2 0 1 0 2 1 0
2 1 k k k k k g k k k k k k g k Y Y y
PR
− − − + − + − − + = − α α (3)
Taking expectations on both sides of (3), we get the bias of
PR
y to the first degree of
approximation, as
( ) ( ) [ ]
2 2
2
2 2
1 2 1
2 2
1
2
) (
z x zx x PR
S R S R S R R S R g gDA Y y B − − − + = θ λ
where θ 2 1− = A
Squaring both sides of (3) and neglecting terms of s k' involving power greater than
two, we have
( ) [ ]
2
2 1 0
2
2
Agk Agk k Y Y y
PR
− + = −
[ ]
2
2
2 2 2
1
2 2
2 1
2 2
2 0 1 0
2
0
2
2 2 2 k g A k g A k k g A k Agk k Agk k Y + + − − + = (4)
48
Taking expectations on both sides of (4), we get the MSE of
PR
y , to the first order of
approximation, as
( ) [ ] C g A AgD S y MSE
y PR
2 2 2
2 + + = λ (5)
The MSE of the proposed estimator given in (5) can be rewritten in terms of
coefficient of variation as
( ) [ ]
∗ ∗
+ + = C g A D AgC C Y y MSE
y y PR
2 2 2 2
2 λ
where
x z zx z x
C C C C C ρ 2
2 2
− + =
∗
and
z yz x yx
C C D ρ ρ − =
∗
,
Y
S
C
y
y
= ,
X
S
C
x
x
= ,
Z
S
C
z
z
=
The MSE equation given in (5) is minimized for
0
2
θ θ =
+
=
Cg
Cg D
(say)
We can obtain the minimum MSE of the suggested estimator
PR
y
,
by using the
optimal equation of θ in (5) as follows: ( ) ( ) [ ] CF D F S y MSE
y PR
+ + = 2 . min
2
λ
where E g F − = and
C
Cg D
E
+
=
3. EFFICIENCY COMPARISON
In this section, the efficiency of the suggested estimator
PR
y
over the following
estimator, y
,
R
y ,
P
y ,
S
y ,
∗
R
y ,
∗
P
y ,
SE
y and
ST
y are investigated. We will have the
conditions as follows:
(a) ( ) ( ) 0 < − y V y MSE
PR
if
gC
gC D
2
2 +
> θ
(b) ( ) ( ) 0 < −
R PR
y MSE y MSE if
( ) ( )
yx x
S S R R AgC D Ag 2 2
2
1 1
− < + provided
2
2
1 x
yx
S R
S <
(c) ( ) ( ) 0 < −
P PR
y MSE y MSE if
49
( ) ( )
yz z
S S R R AgC D Ag 2 2
2
2 2
+ < + provided
2
2
2 z
yz
S R
S <
(d) ( ) ( ) 0 < −
S PR
y MSE y MSE if
( ) 1
2
−
−
<
Ag
D
C provided
A
g
1
<
(e) ( ) ( ) 0 < −
∗
R PR
y MSE y MSE if
( )
2 2
2 1 2 2 1
4 2 2 4
z yx yz zx
S gR S R S R S R gR D gC gC − − + < − − θ θ
(f) ( ) ( ) 0 < −
∗
P PR
y MSE y MSE if
( )
2 2
1 1 2 2 1
2 4 2 4
x yx yz zx
S gR S R S R S R gR D gC gC − − + < − − θ θ
(g) ( ) ( ) 0 < −
SE PR
y MSE y MSE if
( ) 1
2
−
−
<
A C
D
g provided 1 < A
(h) ( ) ( ) 0 < −
ST PR
y MSE y MSE if ( )
2 2
2
y
S AgC D Ag ρ − < +
4. NUMERICAL ILLUSTRATION
To analyze the performance of the suggested estimator in comparison to other estimators
considered in this paper, two natural population data sets from the literature are being
considered. The descriptions of these populations are given below.
(1) Population I [Singh (1969, p. 377]; a detailed description can be seen in Singh (1965)
y : Number of females employed
x : Number of females in service
z : Number of educated females
61 = N , 20 = n , 46 . 7 = Y , 31 . 5 = X , 179 = Z , 0818 . 28
2
=
y
S , 1761 . 16
2
=
x
S ,
1953 . 2028
2
=
z
S , 7737 . 0 =
xy
ρ , 2070 . 0 − =
yz
ρ , 0033 . 0 − =
zx
ρ ,
(2) Population II [Source: Johnston 1972, p. 171]; A detailed description of these
variables is shown in Table 1.
y : Percentage of hives affected by disease
x : Mean January temperature
50
z : Date of flowering of a particular summer species (number of days from January 1)
10 = N , 4 = n , 52 = Y , 42 = X , 200 = Z , 9776 . 65
2
=
y
S , 9880 . 29
2
=
x
S , 84
2
=
z
S ,
8 . 0 =
xy
ρ , 94 . 0 − =
yz
ρ , 073 . 0 − =
zx
ρ ,
For these comparisons, the Percent Relative Efficiencies (PREs) of the different
estimators are computed with respect to the usual unbiased estimator y , using the formula
( )
( )
( )
100
.
., × =
MSE
y V
y PRE
and they are as presented in Table 2.
Table 1: Description of Population II.
y
x z
49 35 200
40 35 212
41 38 211
46 40 212
52 40 203
59 42 194
53 44 194
61 46 188
55 50 196
64 50 190
Table 2 shows clearly that the proposed dual to ratiocumproduct estimator
PR
y
has
the highest PRE than other estimators; therefore, we can conclude based on the study
populations that the suggested estimator is more efficient than the usual unbiased estimators,
the traditional ratio and product estimator, ratiocumproduct estimator by Singh (1969),
51
Srivenkataramana (1980) estimator, Bandyopadhyay (1980) estimator, Singh et al. (2005)
estimator and Sharma and Tailor (2010).
Table 2: PRE of the different estimators with respect to y
Estimators Population I Population II
y
100 100
R
y 205 277
P
y 102 187
S
y 214 395
∗
R
y
215 239
∗
P
y
105 150
SE
y
236 402
ST
y
250 278
PR
y
279 457
5. CONCLUSION
We have developed a new estimator for estimating the finite population mean, which
is found to be more efficient than the usual unbiased estimator, the traditional ratio and
product estimators and the estimators proposed by Singh (1969), Srivenkataramana (1980),
Bandyopadhyay (1980), Singh et al. (2005) and Sharma and Tailor (2010). This theoretical
inference is also satisfied by the result of an application with original data. In future, we hope
to extend the estimators suggested here for the development of a new estimator in stratified
random sampling.
52
REFERENCES
1. ABUDAYYEH, W. A., AHMED, M. S., AHMED, R. A., and MUTTLAK, H. A. (2003):
Some estimators of finite population mean using auxiliary information, Applied Mathematics
and Computation, 139, 287–298.
2. BANDYOPADHYAY, S. (1980): Improved ratio and product estimators, Sankhya series C
42, 45 49.
3. DIANA, G. and PERRI, P.F. (2007): Estimation of finite population mean using multi
auxiliary information. Metron, Vol. LXV, Number 1, 99112
4. JOHNSTON, J. (1972): Econometric methods, (2nd edn), McGrawHill, Tokyo.
5. KADILAR, C. and CINGI, H. (2004): Estimator of a population mean using two auxiliary
variables in simple random sampling, International Mathematical Journal, 5, 357–367.
6. KADILAR, C. and CINGI, H. (2005): A new estimator using two auxiliary variables, Applied
Mathematics and Computation, 162, 901–908.
7. Malik, S. and Singh, R. (2012) : Some improved multivariate ratio type estimators using
geometric and harmonic means in stratified random sampling. ISRN Prob. and Stat. Article
ID 509186, doi:10.5402/2012/509186.
8. MURTHY, M. N. (1964): Product method of estimation. Sankhya, A, 26, 294307.
9. OLKIN, I. (1958): Multivariate ratio estimation for finite populations, Biometrika, 45, 154–
165.
10. PERRI, P. F. (2004): Alcune considerazioni sull’efficienza degli stimatori rapportocum
prodotto, Statistica & pplicazioni, 2, no. 2, 59–75.
11. PERRI, P. F. (2005): Combining two auxiliary variables in ratiocumproduct type estimators.
Proceedings of Italian Statistical Society, Intermediate Meeting on Statistics and
Environment, Messina, 2123 September 2005, 193–196.
12. RAJ, D. (1965): On a method of using multiauxiliary information in sample surveys, Journal
of the American Statistical Association, 60, 154–165.
13. REDDY, V. N. (1974): On a transformed ratio method of estimation, Sankhya C, 36, 59–70.
14. SHARMA, B. and TAILOR, R. (2010): A New RatioCumDual to Ratio Estimator of Finite
Population Mean in Simple Random Sampling. Global Journal of Science Frontier Research,
Vol. 10, Issue 1, 2731
15. SINGH, M.P. (1965): On the estimation of ratio and product of the population parameters,
Sankhya series B 27, 321328.
16. SINGH, M. P. (1967): Multivariate product method of estimation for finite populations,
Journal of the Indian Society of Agricultural Statistics, 31, 375–378.
17. SINGH, M.P. (1969): Comparison of some ratiocumproduct estimators, Sankhya series B
31, 375378.
53
18. SINGH, H.P., SINGH, R., ESPEJO, M.R., PINEDA, M.D and NADARAJAH, S. (2005): On
the efficiency of a dual to ratiocumproduct estimator in sample surveys. Mathematical
Proceedings of the Royal Irish Academy, 105A (2), 5156.
19. SINGH, R., SINGH, H.P AND ESPEJO, M.R. (1998): The efficiency of an alternative to
ratio estimator under a super population model. J.S.P.I., 71, 287301.
20. SINGH, R., CHAUHAN, P. AND SAWAN, N. (2007): On the bias reduction in linear variety
of alternative to ratiocumproduct estimator. Statistics in Transition,8(2),293300.
21. SINGH, R., KUMAR, M., CHAUHAN, P., SAWAN, N. AND SMARANDACHE, F. (2011):
A general family of dual to ratiocumproduct estimator in sample surveys. SIT_newseries,
12(3), 587594.
22. SRIVENKATARAMANA, T. (1978): Change of origin and scale in ratio and difference
methods of estimation in sampling, Canad. J. Stat., 6, 79–86.
23. SRIVENKATARAMANA, T. (1980): A dual to ratio estimator in sample surveys,
Biometrika, 67(1), 199–204.
24. TRACY, D.S., SINGH, H.P. AND SINGH, R. (1996): An alternative to ratiocumproduct
estimator in sample surveys. J.S.P.I.,53,375387.
54
Some Improved Estimators for Population Variance Using Two Auxiliary
Variables in Double Sampling
1
Viplav Kumar Singh, †
1
Rajesh Singh and
2
Florentin Smarandache
1
Department of Statistics, Banaras Hindu University, Varanasi221005, India
2
Department of Mathematics, University of New Mexico, Gallup, USA
†Corresponding author
Abstract
In this article we have proposed an efficient generalised class of estimator using two
auxiliary variables for estimating unknown population variance
2
y
S of study variable y .We
have also extended our problem to the case of two phase sampling. In support of theoretical
results we have included an empirical study.
1. Introduction
Use of auxiliary information improves the precision of the estimate of parameter .Out of
many ratio and product methods of estimation are good example in this context. We can use
ratio method of estimation when correlation coefficient between auxiliary and study variate is
positive (high), on the other hand we use product method of estimation when correlation
coefficient between auxiliary and study variate is highly negative.
Variations are present everywhere in our daytoday life. An agriculturist
needs an adequate understanding of the variations in climatic factors especially from place to
place (or time to time) to be able to plan on when, how and where to plant his crop. The
problem of estimation of finite population variance
2
y
S , of the study variable y was discussed
by Isaki (1983), Singh and Singh (2001, 2002, 2003), Singh et al. (2008), Grover (2010), and
Singh et al. (2011).
55
Let x and z are auxiliary variates having values ) z , x (
i i
and y is the study variate having
values ) y (
i
respectively. Let ) N ,....... 2 , 1 i ( V
i
= is the population having N units such that y is
positively correlated x and negatively correlated with z. To estimate
2
y
S , we assume that
2
z
2
x
S and S are known, where
¿
=
− =
N
1 i
2
i
2
y
) Y y (
N
1
S ,
¿
=
− =
N
1 i
2
i
2
x
) X x (
N
1
S and . ) Z z (
N
1
S
N
1 i
2
i
2
z ¿
=
− =
Assume that N is large so that the finite population correction terms are ignored. A sample of
size n is drawn from the population V using simple random sample without replacement.
Usual unbiased estimator of population variance
2
y
S is , s
2
y
where, . ) y y (
) 1 n (
1
s
n
1 i
2
i
2
y ¿
=
−
−
=
Up to the first order of approximation, variance of
2
y
s is given by
*
400
4
y 2
y
n
S
) s var( ∂ = (1.1)
where, 1
400
*
400
− ∂ = ∂ ,
2 / r
002
2 / q
020
2 / p
200
pqr
pqr
μ μ μ
μ
= ∂ , and
¿
=
− − − = μ
N
1 i
r
i
q
i
p
i pqr
) Z z ( ) X x ( ) Y y (
N
1
; p, q, r being the nonnegative integers.
2. Existing Estimators
Let ) e 1 ( S s and ) e 1 ( S s , ) e 1 ( S s
2
2
z
2
z 1
2
x
2
x 0
2
y
2
y
+ = + = + =
where,
¿ ¿
= =
−
−
= −
−
=
n
1 i
n
1 i
2
i
2
z
2
i
2
x
) z z (
) 1 n (
1
s , ) x x (
) 1 n (
1
s
and . z
n
1
z , x
n
1
x
n
1 i
i
n
1 i
i ¿ ¿
= =
= =
Also, let
0 ) e ( E ) e ( E
2 1
= =
56
n
) e ( E and
n
) e ( E ,
n
) e ( E
*
004 2
2
*
040 2
1
*
400 2
0
∂
=
∂
=
∂
=
*
202 2 0
*
022 2 1
*
220 1 0
n
1
) e e ( E ,
n
1
) e e ( E ,
n
1
) e e ( E ∂ = ∂ = ∂ =
Isaki (1983) suggested ratio estimator
1
t for estimating
2
y
S as
2
x
2
y 2
y 1
s
S
s t = ; where
2
x
s is unbiased estimator of
2
x
S (1.2)
Up to the first order of approximation, mean square error of
1
t is given by,
[ ]
*
220
*
040
*
400
4
y
1
2
n
S
) t ( MSE ∂ − ∂ + ∂ = (1.3)
Singh et al. (2007) proposed the exponential ratiotype estimator
2
t as
(
¸
(
¸
+
−
=
2
x
2
x
2
x
2
x 2
y 2
s S
s S
exp s t (1.4)
And exponential product type estimator
3
t as
(
¸
(
¸
+
−
=
2
x
2
x
2
x
2
x 2
y 3
S s
S s
s t (1.5)
Following Kadilar and Cingi (2006), Singh et al. (2011) proposed an improved estimator for
estimating population variance
2
y
S , as
(
(
¸
(
¸
)
`
¹
¹
´
¦
+
−
− +
)
`
¹
¹
´
¦
+
−
=
2
x
2
x
2
x
2
x
4
2
x
2
x
2
x
2
x
4
2
y 4
S s
S s
) k 1 (
s S
s S
exp k s t (1.6)
where
4
k is a constant.
Up to the first order of approximation mean square errors of
2
t ,
3
t and
4
t are respectively
given by
(
¸
(
¸
∂ −
∂
+ ∂ =
*
220
*
040 *
400
4
y
2
4 n
S
) t ( MSE (1.7)
57
(
¸
(
¸
∂ −
∂
+ ∂ =
*
202
*
004 *
400
4
y
3
4 n
S
) t ( MSE (1.8)
(
¸
(
¸
∂
−
− ∂ − + ∂ −
∂
− +
∂
+ ∂ =
*
022
4 4 *
202 4
*
220 4
*
004 2
4
*
040 2
4
*
400
4
y
4
2
) k 1 ( k
) k 1 ( k
4
) k 1 (
4
k
n
S
) t ( MSE (1.9)
where .
( 2
) 2 / (
k
*
022
*
004
*
040
*
022
*
220
*
004
4
∂ + ∂ + ∂
∂ + ∂ + ∂
=
3. Improved Estimator
Using Singh and Solanki (2011), we propose some improved estimators for estimating
population variance
2
y
S as
p
2
x
2
x
2
x 2
y 5
S ) d c (
Ds cS
s t
(
¸
(
¸
−
−
= (1.10)
q
2
x
2
x
2
z 2
y 6
bs aS
S ) b a (
s t
(
¸
(
¸
+
+
= (1.11)
(
(
¸
(
¸
)
`
¹
¹
´
¦
+
+
− +
)
`
¹
¹
´
¦
−
−
=
q
2
x
2
x
2
z
7
p
2
x
2
x
2
x
7
2
y 7
bs aS
S ) b a (
) k 1 (
S ) d c (
Ds cS
k s t (1.12)
where a, b, c, d are suitably chosen constants and
7
k is a real constant to be determined so
as to minimize MSE’s.
Expressing
5
t ,
6
t and
7
t in terms of s ' e
i
, we have
[ ]
1 0 1 1 1
2
y 5
e pe x e pe x 1 S t
0
− + − = (1.13)
where, .
) d c (
d
x
1
−
=
[ ]
2 0 2 0 2 2
2
y 6
e e qx e e qx 1 S t − + − = (1.14)
where, .
) b a (
b
x
2
+
=
[ ]
'
2 2 7 1
'
1 7 1 0
2
y 7
e qx ) 1 k ( ) e e ( k px e 1 S t − + − + + = (1.15)
58
The mean squared error of estimators are obtained by subtracting
2
y
S from each estimator and
squaring both sides and than taking expectations
[ ]
*
220 1
*
040
2 2
1
*
400
4
y
5
p x 2 p x
n
S
) t ( MSE ∂ − ∂ + ∂ = (1.16)
Differentiating (1.16) with respect to
1
x , we get the optimum value of
1
x as
.
p
x
*
040
*
220
) opt ( 1
∂
∂
=
[ ]
*
202 1
*
004
2 2
2
*
400
4
y
6
x 2 q x
n
S
) t ( MSE ∂ − ∂ + ∂ = (1.17)
Differentiating (1.17) with respect to
2
x , we get the optimum value of
2
x as –
.
q
x
*
004
*
202
) opt ( 2
∂
∂
=
[ ] F ) k 1 ( 2 E ) k 1 ( k 2 C ) k 1 ( B k A
n
S
) t ( MSE
7 7 7 7
2
7
4
y
7
− − − − − + + = (1.18)
Differentiating (1.18) with respect to
7
k , we get the optimum value
7
k of as –
.
E 2 C B
E F D C
k
) opt ( 7
− +
− − +
=
where
*
400
A ∂ = ,
*
040
2 2
1
p x B ∂ = ,
*
004
2 2
2
q x C ∂ = , , p x D
*
220 1
∂ =
*
022 2 1
pq x x E ∂ = , . q x F
*
202 2
∂ =
2. Estimators In Two Phase Sampling
In certain practical situations when
2
x
S is not known a priori, the technique of two phase
sampling or double sampling is used. Allowing SRSWOR design in each phase, the two –
phase sampling scheme is as follows:
The first phase sample ) V s ( s
'
n
'
n
⊂ of a fixed size n’ is drawn to measure only x and z
in order to formulate the a good estimate of
2
x
S and
2
z
S , respectively.
59
Given
'
n
s ,the second phase sample ) s s ( s
'
n n n
⊂ of a fixed size n is drawn to measure y
only.
Existing Estimators
Singh et al. (2007) proposed some estimators to estimate
2
y
S in two phase sampling, as:
(
¸
(
¸
+
−
=
2
x
2
x
2
x
2
x 2
y
'
2
s ' s
s ' s
exp s t (2.1)
(
¸
(
¸
+
−
=
2
z
2
z
2
z
2
z 2
y
'
3
s ' s
s ' s
exp s t (2.2)
(
(
¸
(
¸
)
`
¹
¹
´
¦
+
−
− +
)
`
¹
¹
´
¦
+
−
=
2
z
2
z
2
z
2
z '
4
2
x
2
x
2
x
2
x '
4
2
y
'
4
s ' s
s ' s
exp ) k 1 (
s ' s
s ' s
exp k s t (2.3)
MSE of the estimator
'
2
t ,
'
3
t and
'
4
t are respectively, given by
(
¸
(
¸
∂ 
.

\

− + ∂ 
.

\

− +
∂
=
*
220
*
040
*
400 4
y
'
2
n
1
' n
1
' n
1
n
1
4
1
n
S ) t ( MSE (2.4)
(
¸
(
¸
∂ 
.

\

− − ∂ 
.

\

− +
∂
=
*
202
*
004
*
400 4
y
'
3
n
1
' n
1
' n
1
n
1
4
1
n
S ) t ( MSE (2.5)
[ ] ' E ) k 1 ( ' D k ' C ) k 1 ( ' B ' k ' A S ) t ( MSE
'
4
'
4
2 '
4
2
4
4
y
'
4
− + + − + + = (2.6)
And
) ' C ' B ( 2
' D ' E ' C 2
k
) opt (
'
4
+
− +
=
Where,
*
202
*
202
*
004
*
040
*
400
' n
1
' E ,
n
1
' n
1
' D
' n 4
1
' C
' n
1
n
1
4
1
' B ,
n
' A
∂ = ∂ 
.

\

− =
∂ = ∂ 
.

\

− =
∂
=
Proposed estimators in two phase sampling
The estimator proposed in section 3 will take the following form in two phase sampling;
(
¸
(
¸
−
−
=
2
x
2
x
2
x 2
y
'
5
' s ) d c (
ds ' cs
s t (2.7)
60
(
¸
(
¸
+
+
=
2
z
2
z
2
z 2
y
'
6
' bs aS
S ) b a (
s t (2.8)
(
(
¸
(
¸
)
`
¹
¹
´
¦
+
+
− +
)
`
¹
¹
´
¦
−
−
=
2
z
2
z
2
z '
7
2
x
2
x
2
x '
7
2
y
'
7
' bs aS
S ) b a (
) k 1 (
' s ) d c (
ds ' cs
k s t (2.9)
Let,
) e 1 ( S ' s ), e 1 ( S s , ) e 1 ( S s
'
1
2
x
2
x 1
2
x
2
x 0
2
y
2
y
+ = + = + =
) e 1 ( S ' s , ) e 1 ( S s
'
2
2
z
2
z 2
2
z
2
z
+ = + =
Where,
¿ ¿
¿ ¿
= =
= =
= =
−
−
= −
−
=
' n
1 i
i
' n
1 i
i
' n
1 i
2
i
2
z
' n
1 i
2
i
2
x
z
' n
1
' z , x
' n
1
' x and
) ' z z (
) 1 ' n (
1
' s , ) ' x x (
) 1 ' n (
1
' s
Also,
*
022
'
2 1
*
022
'
2 1
*
004
'
2 2
*
040
'
1 1
*
202
'
2 0
*
220
'
1 0
*
004
2
'
2
*
040
2
'
1
'
2
'
1
' n
1
) e ' e ( E ,
' n
1
) e e ( E ,
' n
1
) e e ( E
' n
1
) e e ( E ,
' n
1
) e e ( E ,
' n
1
) e e ( E
' n
) e ( E ,
' n
) e ( E
0 ) e ( E ) e ( E
∂ = ∂ = ∂ =
∂ = ∂ = ∂ =
∂
=
∂
=
= =
Writing estimators
'
5
t ,
'
6
t and
'
7
t in terms of s ' e
i
we have ,respectively
[ ] ) e e ( px e 1 S t
1
'
1 1 0
2
y
'
5
− + + = (2.10)
[ ]
'
2 2 0
2
y
'
6
e qx e 1 S t − + = (2.11)
[ ]
'
2 2
'
7 0 1
'
1
'
7 1
2
y
'
7
e qx ) 1 k ( e ) e e (( k px 1 S t − + + − + = (2.12)
Solving (1.10),(1.11) and (1.12),we get the MSE’S of the estimators
'
5
t ,
'
6
t and
'
7
t ,
respectively as
(
¸
(
¸
∂ 
.

\

− + ∂ 
.

\

− +
∂
=
*
220 1
*
040
2
1
2
*
400 4
y
'
5
n
1
' n
1
px 2
' n
1
n
1
x p
n
S ) t ( MSE (2.13)
Differentiate (2.13) w.r.t.
1
x , we get the optimum value of
1
x as
61
*
040
*
220
) opt ( 1
p
x
∂
∂
=
(
¸
(
¸
∂ + ∂ +
∂
=
*
202 2
*
004
2
2
2
*
400 4
y
'
6
' n
1
qx 2
' n
1
x q
n
S ) t ( MSE (2.14)
Differentiate (2.14) with respect to
2
x , we get the optimum value of
2
x as –
*
004
*
202
) opt ( 2
q
x
∂
∂
= .
[ ]
1
'
7 1
'
7 1
2 '
7
'
7 1 1
4
y
'
7
E ) 1 k ( 2 D k 2 C ) 1 k ( k B A S ) t ( MSE
2
− + + − + + = (2.15)
Differentiate (2.15) with respect to
'
7
k , we get the optimum value of
'
7
k as –
1 1
1 1 1
) opt (
'
7
C B
E D C
k
+
− −
=
Where,
n
A
*
400
1
∂
= ,
*
040
2
1
2
1
' n
1
n
1
x p B ∂ 
.

\

− = ,
*
004
2
2
2
1 1
' n
x q
px C ∂ =
*
220 1 1
n
1
' n
1
px D ∂ 
.

\

− = ,
' n
qx E
*
202
2 1
∂
=
5. Empirical Study
In support of theoretical result an empirical study is carried out. The data is taken from
Murthy(1967):
808 . 2 , 912 . 2 , 726 . 3
044 040 400
= ∂ = ∂ = ∂
105 . 3 , 979 . 2 , 73 . 2
220 202 022
= ∂ = ∂ = ∂
7205 . 0 c , 7531 . 0 c , 5938 . 0 c
z y x
= = =
15 ' n , 7 n , 98 . 0 , 904 . 0
xy yz
= = = ρ = ρ
8824 . 208 Z , 4412 . 199 Y , 5882 . 747 X = = =
62
In Table 5.1 percent relative efficiency of various estimators of
2
y
S is written with
respect to
2
y
s
Table 5.1: PRE of the estimator with respect to
2
y
s
Estimators PRE
2
y
s
100
1
t
636.9158
2
t
248.0436
3
t
52.86019
4
t
699.2526
5
t
667.2895
6
t
486.9362
7
t
699.5512
63
In Table 5.2 percent relative efficiency of various estimators of
2
y
S is written with
respect to
2
y
s in two phase sampling:
Table 5.2: PRE of the estimators in two phase sampling with respect to
2
y
s
6. Conclusion
In Table 5.1 and 5.2 percent relative efficiencies of various estimators are
written with respect to . s
2
y
From Table 5.1 and 5.2 we observe that the proposed estimator
Estimators PRE
2
y
s
100
' t
2
142.60
' t
3
66.42
' t
4
460.75
' t
5
182.95
' t
6
158.93
' t
7
568.75
64
under optimum condition performs better than usual estimator, Isaki (1983) estimator and
Singh et al. (2007 ) estimator.
7. References
Bahl, S. and Tuteja, R.K. (1991): Ratio and Product type exponential estimator. Information and
Optimization sciences, Vol. XII, I, 159163.
Grover, L. K . (2010): A correction note on improvement in variance estimation using auxiliary
information. Commun. Stat. Theo. Meth. 39:753–764.
Isaki, C. T. (1983): Variance estimation using auxiliary information. Journal of American Statistical
Association.
Kadilar,C. and Cingi,H. (2006) : Improvement in estimating the population mean in simple
random sampling. Applied Mathematics Letters 19 (2006), 75–79.
Murthy, M. N. (1967).: Sampling Theory and Methods, Statistical Publishing Soc., Calcutta, India.
Singh, H. P. and Singh, R. (2001): Improved Ratiotype Estimator for Variance Using Auxiliary
Information. J.I.S.A.S.,54(3),276287.
Singh, H. P. and Singh, R. (2002): A class of chain ratiotype estimators for the coefficient of
variation of finite population in two phase sampling. Aligarh Journal of Statistics, 22,19.
Singh, H. P. and Singh, R. (2003): Estimation of variance through regression approach in two phase
sampling. Aligarh Journal of Statistics, 23, 1330.
Singh, H. P, and Solanki, R. S. (2012): A new procedure for variance estimation in simple random
sampling using auxiliary information. Statistical papers, DOI 10.1007/s0036201204452.
Singh, R., Chauhan, P., Sawan, N. and Smarandache, F. (2011): Improved exponential estimator for
population variance using two auxiliary variables. Italian Jour. Of Pure and Applied. Math., 28, 103
110.
Singh, R., Kumar, M., Singh, A.K. and Smarandache, F. (2011): A family of estimators of
population variance using information on auxiliary attribute. Studies in sampling techniques and
time series analysis. Zip publishing, USA.
Singh, R., Chauhan, P., Sawan, N. and Smarandache, F. (2008): Almost unbiased ratio and product
type estimator of finite population variance using the knowledge of kurtosis of an auxiliary variable in
sample surveys. Octogon Mathematical Journal, Vol. 16, No. 1, 123130.
The purpose of writing this book is to suggest some improved estimators using
auxiliary information in sampling schemes like simple random sampling and systematic
sampling.
This volume is a collection of five papers, written by eight coauthors (listed in the
order of the papers): Manoj K. Chaudhary, Sachin Malik, Rajesh Singh, Florentin
Smarandache, Hemant Verma, Prayas Sharma, Olufadi Yunusa, and Viplav Kumar
Singh, from India, Nigeria, and USA.
The following problems have been discussed in the book:
In chapter one an estimator in systematic sampling using auxiliary information
is studied in the presence of nonresponse. In second chapter some improved estimators
are suggested using auxiliary information. In third chapter some improved ratiotype
estimators are suggested and their properties are studied under second order of
approximation.
In chapter four and five some estimators are proposed for estimating unknown
population parameter(s) and their properties are studied.
This book will be helpful for the researchers and students who are working in
the field of finite population estimation.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue listening from where you left off, or restart the preview.