Professional Documents
Culture Documents
Some Ratio and Product Estimators Using Known Value of Population Parameters
Some Ratio and Product Estimators Using Known Value of Population Parameters
Some Ratio and Product Estimators Using Known Value of Population Parameters
Florentin Smarandache
Department of Mathematics, University of New Mexico
Gallup, NM 87301, USA
Published in:
Sachin Malik, Neeraj Kumar, Florentin Smarandache (Editors)
USES OF SAMPLING TECHNIQUES & INVENTORY CONTROL
WITH CAPACITY CONSTRAINTS
Pons Editions, Brussels, Belgium, 2016
ISBN 978-1-59973-484-2
pp. 19 - 29
Abstract
Keywords: Ratio Estimator, Product Estimator, Population Parameter, Efficiency, Mean Square
Error.
1. Introduction
The problem of estimating the population mean in the presence of an auxiliary variable
has been widely discussed in finite population sampling literature. Ratio, product and difference
methods of estimation are good examples in this context. Ratio method of estimation is quite
effective when there is high positive correlation between study and auxiliary variables. On the
other hand, if correlation is negative (high), the product method of estimation can be employed
efficiently.
19
Uses of Sampling Techniques & Inventory Control with Capacity Constraints
aX b
t y[ ]g
α(ax b) (1 α)(aX b)
where a( 0) , b are either real numbers or the functions of the known parameters of the auxiliary
variable x such as standard deviation ( σ x ), Coefficient of Variation ( C x ), Skewness ( β1 (x) ),
Kurtosis ( β 2 (x) ) and Correlation Coefficient ( ρ ).
(i). When α=0, a=0=b, g=0, we have the mean per unit estimator, t 0 y with
Nn 2 2
MSE(t 0 ) ( )Y C y (1.1)
Nn
X
(ii). When α=1, a=1, b=0, g=1, we have the usual ratio estimator, t 1 y( ) with
x
Nn 2
MSE(t 1 ) ( )Y (C y C x 2ρρ x C y )
2 2
(1.2)
Nn
x
(iii). When α=1, a=1, b=0, g=-1, we have the usual product estimator, t 2 y( ) with
X
Nn 2
MSE(t 2 ) ( )Y (C y C x 2ρρ x C y )
2 2
(1.3)
Nn
(iv). When α=1, a=1, b= C x , g=1, we have Sisodia and Dwivedi [2] ratio estimator,
X Cx
t 3 y( ) with
x Cx
Nn 2 X X
MSE(t 3 ) ( )Y (C y ( ) 2 C x 2(
2 2
)ρρ x C y ) (1.4)
Nn X Cx X Cx
20
Some Ratio and Product Estimators Using Known Value of Population Parameters
x Cx
we have Pandey and Dubey [4] product estimator, t 4 y( ) with
X Cx
Nn 2 X X
MSE(t 4 ) ( )Y (C y ( ) 2 C x 2(
2 2
)ρρ x C y ) (1.5)
Nn X Cx X Cx
Xρ
(vi). When α=1, a=1, b= , g=1, we have Singh and Taylor [3] ratio estimator, t 5 y( )
xρ
with
Nn 2 X 2 2 X
MSE(t 5 ) ( )Y (C y ( ) C x 2(
2
)ρρ x C y ) (1.6)
Nn Xρ Xρ
(vii). When α=1, a=1, b= ρ , g=-1, we have Singh and Taylor [3] product estimator,
xρ
t 6 y( ) with
Xρ
Nn 2 X 2 2 X
MSE(t 6 ) ( )Y (C y ( ) C x 2(
2
)ρρ x C y ) (1.7)
Nn Xρ Xρ
There are other ratio and product estimators from these families that are not inferred here but this
paper will be limited to those ones that made use of Coefficient of Variation ( Cx ) and
Correlation Coefficient ( ρ ) since the conclusion obtained here can also be inferred on all others
that made use of other population parameters such as the standard deviation ( σ x ), Skewness
( β1 (x) ) and Kurtosis ( β 2 (x) ) in the same family.
Adopting Adewara (2006), Adewara et al. (2012) proposed the following estimators as
X
t *1 y * ( ), (2.1)
x*
x*
t *2 y* ( ), (2.2)
X
X Cx
t *3 y* ( ), (2.3)
x* Cx
21
Uses of Sampling Techniques & Inventory Control with Capacity Constraints
x* Cx
t *
4 y (
*
), (2.4)
X Cx
Xρ
t *5 y* ( ) and (2.5)
x* ρ
x* ρ
t *6 y* ( ) , (2.6)
Xρ
Where x * and y * are the sample means of the auxiliary variables and variable of interest yet to
n 2
(i). MSE(t *1 ) ( ) MSE(t 1 ) (2.7)
Nn
n 2
(ii). MSE(t * 2 ) ( ) MSE(t 2 ) (2.8)
Nn
n 2
(iii). MSE(t * 3 ) ( ) MSE(t 3 ) (2.9)
Nn
n 2
(iv). MSE(t * 4 ) ( ) MSE(t 4 ) (2.10)
Nn
n 2
(v). MSE(t * 5 ) ( ) MSE(t 5 ) (2.11)
Nn
n 2
(vi). MSE(t * 6 ) ( ) MSE(t 6 ) (2.12)
Nn
Following Adewara et al [5], Yadav and Kadilar [6] proposed some improved ratio and product
estimators for estimating the population mean of the study variable as follows
X
η*1 ky * ( ), (2.13)
x*
x*
η* 2 k y * ( ), (2.14)
X
22
Some Ratio and Product Estimators Using Known Value of Population Parameters
X Cx
η* 3 k y * ( ), (2.15)
x* Cx
x* Cx
η *
4 ky ( *
), (2.16)
X Cx
Xρ
η* 5 k y * ( ) (2.17)
x* ρ
x* ρ
η *
6 ky ( *
) , (2.18)
Xρ
MSE(η*1 ) Y h 2 k 12 λC 2y 3k12 2k1 λC 2x 2 2k12 k 1 λC yx k 1 1
2
2
(2.19)
MSE(η* 2 ) Y h 2 k 22 λC 2y k 22 λC 2x 2 2k12 k 1 λC yx k 2 1
2
2
(2.20)
MSE(η* 3 ) Y h 2 k 32 λC 2y 3k 32 2k 3 ν 12 λC 2x 2ν 1 2k 32 k 3 λC yx k 3 1
2
2
(2.21)
MSE(η* 4 ) Y h 2 k 24 λC 2y k 24 ν 12 λC 2x 2ν 1 2k 24 k 4 λC yx k 4 1
2
2
(2.22)
MSE(η* 3 ) Y h 2 k 32 λC 2y 3k 32 2k 3 ν 22 λC 2x 2ν 2 2k 52 k 5 λC yx k 5 1
2
2
(2.23)
MSE(η* 4 ) Y h 2 k 62 λC 2y k 24 ν 22 λC 2x 2ν 2 2k 62 k 6 λC yx k 6 1
2
2
(2.24)
Where,
N-n n S2y 2 S2x S yx X X S yx
λ ,h , C y 2 , C x 2 , C yx
2
, ν1 , ν1 andρ
Nn N-n Y X YX X Cx X Cx S ySx
And k 1
h 2 λC 2x λC yx 1
,k2
h 2 λC yx 1
h 2 3C2x λ 4C yx λ λC 2y 1
h 2 C 2x λ 4C yx λ λC 2y 1 ,
k3
h 2 λν 12 C 2x ν 1 λC yx 1 ,k4
h 2 λν 1 λC yx 1
h 3ν C λ 4ν 1C yx λ λC 1
2 2
1
2
x
2
y 2
h 3ν C λ 4ν 1C yx λ λC 2y 1
2
1
2
x
k5
h 2 λν 22 C 2x ν 2 λC yx 1 , andk 6
h 2 λν 2 λC yx 1
h 2 3ν 22 C 2x λ 4ν 2 C yx λ λC 2y 1
h 2 3ν 22 C 2x λ 4ν 2 C yx λ λC 2y 1
23
Uses of Sampling Techniques & Inventory Control with Capacity Constraints
Following Malik Singh [14], we define the following class of estimators for population mean Y
as
t M m1 y m 2 X x *
* *
ψX δ ωX μ ωx * μ
exp
α
β
ψx δ ωX μ ωx μ
*
(3.1)
Where m1 and m 2 are suitably chosen constants. ψ, δ, ω , and μ are either real numbers or
function of known parameters of the auxiliary variable. The scalar α and β takes values +1 and -1
for ratio and product type estimators respectively.
y Y1 e 0 , x X1 e1
expressing equation (3.1) in terms of e’s and retaining only terms up to second degree of e’s, we
have
α β
t M m1 Y1 he 0 m 2 Xhe1
ψX δ
exp
ωXhe1
ψX1 he1 δ 2ωX 2μ - ωXhe1
R he R 2 h 2 e 2
m1 Y1 he 0 m 2 Xhe1 1 R 1he1 expβR 2 he1 1 2 1 2 1
-α
2 4
βh 2 e12
m 2 X he1 αh 2 e12
2 (3.2)
24
Some Ratio and Product Estimators Using Known Value of Population Parameters
ψX ωX
R1 , R2
where, ψX δ ωX μ
t M
Y m1 Y 1 - he 0 L1e1 L 2 e12 L3e 0 e1 m 2 X he1 L 4 e12 Y (3.3)
βhR 2
L1 αR 1h
2
αα 1h R 1 αβh 2 R 1R 2 β 2 h 2 R 22 βh 2 R 22
2 2
L2
2 2 8 4
where,
βh R 2
2
L 3 αh 2 R 1
2
βh 2 R 2
L 4 αh R 1
2
Squaring both sides of (3.3) and neglecting terms of e’s having power greater than two, we have
MSEt M Y 1 m12 T1 m 22 T2 2m1m 2 T3 2m1T4 2m2 T5
2
(3.4)
where,
T1 Y 1 λh 2 C 2y L21λC 2x 2hL1λρC y C x 2L2 λC 2x 2L3 λρC y C x
2
2 2
T2 h λ X C x
2
T3 Y X L 4 λC 2x L1λhC 2x h 2 λρC y C x
2
T4 Y 1 L 2 λC 2x L 3 λρC y C x
T5 Y XL 4 λC x
2
m1
T2 T4 T3T 5 , m2
T1T5 T3T 4
T1T2 T 2
3 T1T2 T32
4. Empirical Study:
25
Uses of Sampling Techniques & Inventory Control with Capacity Constraints
4. Results:
Table 4.1: Showing the estimates obtained for both the Khoshnevisan et al. [1] estimators and
Adewara et al. [5] estimators
t0 5411349 0.5676
t1 2542740 -
t2 - 0.3387
t3 2542893 -
t4 - 0.3388
t5 2542803 -
t6 - 0.3376
t *1 137519.8 -
t *2 - 0.03763
t *3 137528 -
t *4 - 0.03765
t *5 137523.1 -
- 0.03751
t *6
26
Some Ratio and Product Estimators Using Known Value of Population Parameters
Table 4.2: Showing the estimates obtained for Yadav and Kadilar [6] estimators
η*1 136145.37 -
η* 2 - 0.03762
η* 3 136138.05 -
η* 4 - 0.03764
η* 5 136107.94 -
η* 6 - 0.03750
MSE
1 0 1 0 1 0 - - t *1 137519.8 -
1 0 -1 0 1 0 - - t *2 - 0.03763
1 0 1 0 1 Cx - - t *3 137528 -
1 0 -1 0 1 Cx - - t *4 - 0.03765
1 0 1 0 1 ρ - - t *5 137523.1 -
1 0 -1 0 1 ρ - - t *6 - 0.03751
m1 0 1 0 1 0 - - η*1 136145.37 -
m1 0 -1 0 1 0 - - η* 2 - 0.03762
27
Uses of Sampling Techniques & Inventory Control with Capacity Constraints
m1 0 1 0 1 Cx - - η* 3 136138.05 -
m1 0 -1 0 1 Cx - - η* 4 - 0.03764
m1 0 1 0 1 ρ - - η* 5 136107.94 -
m1 0 -1 0 1 ρ - - η* 6 - 0.03750
m1 m2 1 1 1 1 1 1 tM 75502.23 -
m1 m2 -1 -1 1 1 1 1 tM - 0.03370
Since conventionally, for ratio estimators to hold, ρ 0 and also for product estimators to
hold, ρ 0 . Therefore two data sets are used in this paper, one to determine the efficiency of the
modified ratio estimators and the other to determine that of the product estimators as stated
below.
5. Conclusion
In this paper, we have proposed a new family of estimator for estimating unknown
population mean of study variable using auxiliary variable. Expressions for the MSE of the
estimator are derived up to first order of approximation. The proposed family of estimator is
compared with the several existing estimators in literature. From table 4.3, we observe that the
new family of estimators performs better than the other estimators considered in this paper for
both of the data sets.
References
[1] 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 Jour. of theor. statist. 22(2), 181 – 191.
[2] Sisodia, B.V.S. and Dwivedi, V.K.(1981): A modified ratio estimator using coefficient of variation of
auxiliary variable. Journal Ind. Soc. Agril. Statist.,33(2), 13 – 18.
[3] Singh, H.P. and Tailor, R. (2003) : Use of known correlation coefficient in estimating the finite
population mean. Statist. in Trans. 6(4). 555 – 560.
28
Some Ratio and Product Estimators Using Known Value of Population Parameters
[4] Pandey, B.N. and Dubey, Vyas (1988): Modified product estimator using coefficient of variation of
auxiliary variate, Assam Statistical Rev., 2(2), 64 – 66
[5] A.A. Adewara, R. Singh, M. Kumar(2012) : Efficiency of some modified ratio and product estimators
using known value of some population parameters, International Journal of Applied
Science and Technology 2 (2) 76–79.
[6] Yadav, S.K. and Kadilar, C. (2013): Improved class of ratio and product estimators. Applied
mathematics and computation.
[7] Shabbir, J., Gupta, S. (2005). : Improved ratio estimators in stratified sampling. Amer. J. Mathemat.
Manag. Sci. 25:293–311
[8] Shabbir, J. and Gupta, S. (2006) : A new estimator of population mean in stratified sampling,
Commun. Stat. Theo. Meth. 35: 1201–1209
[9] Kadilar, C. and Cingi, H. (2004) : Ratio Estimators in Simple Random Sampling. Appl. Mathe. and
Comput. 151, 893-902
[10] Kadilar, C. and Cingi, H. (2006).: Ratio estimators for the population variance in sample and
stratified random sampling. Appl. Mathe. and Comput. 173, 1047 – 1059.
[11] Adewara, A.A. (2006). : Effects of improving both the auxiliary and variable of interest in ratio and
product estimators. Proc. Pakistan Acad. Sci.43(4): 275 – 278.
[12] Srivenkataramana, T. and Srinath, K.P. (1976) : Ratio and Product methods of estimation in sample
surveys when the two variables are moderately correlated. Vignana Bharathi 2: 54 – 58
[13] Maddala, G.S. (1977): Econometrics. “McGraw Hills Pub. Co.” New York .
[14] Malik and Singh (2015): Estimation of population mean using information on auxiliary attribute in
two-phase sampling. Applied Mathematics and computation, 261 (2015) 114-1
29