International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp.

612-627
© Research India Publications. http://www.ripublication.com

A Modified Class of Log-Type Estimators for Population Mean Using

Auxiliary Information on Variables

Shashi Bhushan1, Raksoni Gupta2, Saurabh Singh1 and Anoop Kumar1

1.
Department of Mathematics and Statistics, Dr. Shakuntala Misra National Rehabilitation University, Lucknow, U.P, India.
2.
Department of Statistics, National Post Graduate College,
Lucknow, U.P., India.

Abstract have proposed an alternative class of estimators which

includes the estimators / classes of estimators given by,
In this paper, a generalised class of estimators is proposed for
Khoshnevisan et al (2007), Koyuncu and Kalidar (2009a),
the estimation of population mean y using the auxiliary Bhushan et al (2015) and Bhushan et al (2015a). It has been
variable. Bias and MSE of the proposed class of estimator is shown that the suggested class of estimator is more efficient
obtained upto the first order of approximation. It has been among all the estimators discussed within, in the sense of
proven that the proposed class of estimators are more efficient lesser mean square error, both theoretically and numerically.
than the usual regression, ratio, and product type estimators.
This fact is also supported through an empirical study which Consider a finite population U  U1 ,U 2 ,U N  of size N
is given at the end as an illustration. from which a sample of size n is drawn according to simple
Key words: study variable, auxiliary variable, bias, mean random sampling without replacement (SRSWOR). Let yi
square error, efficiency. and xi denotes the values of the study and auxiliary variables
for the i th unit,  i  1, 2,..., N  of the population. Further let
1. INTRODUCTION y and x be the sample means of the study and auxiliary

Using auxiliary information is a famous trend in survey variables, respectively. Let us define e0  y  Y   Y and

 
sampling as it increases the efficiency of the estimators.
e1  x  X X .
Sampling literature deals with variety of such type of
estimators using the auxiliary information. When the
correlation coefficient between the study variable and
auxiliary variable is positive , we use ratio estimator whereas Using these notations,

  E  e12   C y2 ,
if the correlation is negative, product estimator is highly
E  e0   E  e1   0 , E e02  Cx2 ,
preferable for better estimation of the population mean.
Dwivedi (1981), Pandey and Dubey (1988), Upadhyaya and E  e0 e1   Cxy  Cx Cy   kCx2
Singh (1999), Singh and Tailor (2003), Singh (2003), Kalidar
and Cingi (2006), Singh et al. (2004, 2008), Bhushan et al. where
(2017a, 2017b, 2017c), Bhushan and Misra (2017d), Bhushan
2 2
and Gupta(2019a,2019b, 2019c, 2019d, 2020) and Singh and Cx2  S x2 X , C y2  S y2 Y , Cxy  S xy X Y ,
Agnihotri (2008) have also made use of various parameters of
   
N N
S x2   xi  X  N  1 , S y2   yi  Y  N  1 ,
2 2
the auxiliary variable such as standard deviation, coefficient
of variation, correlation coefficient, coefficient of skewness, i 1 i 1

coefficient of kurtosis, etc. The efficiency of the estimators

  yi  Y  k    Cy Cx  and
N
proposed by the above authors is approximately equal to the S xy   xi  X  N  1 ,
i 1
usual regression estimator. Recently Bhushan et al (2015)
have proposed a generalised class of logarithmic type    N  n  Nn
estimators, probably for the first time, for estimating
population mean. They have also given a better class of log
type estimators using the Searls approach. In this paper, we

612
 International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

2. THE SUGGESTED CLASS OF ESTIMATORS (iii) For  w1 , w2    0,1 , the class of estimators T can be
reduced to the class of estimators given by Bhushan et al
We define a class of estimators for the population mean Y as (2015)

  x*  
g
   
 x*  
*
X 
T  w1 y  *   w2 y 1  log    (2.1) 
  y 1  log   
  x  1    X *    X*     X*  
(2.5)
        

where
(iv) For  w1 , w2    0, w2  , the class of estimators T
*
X  aX b transforms to the class of estimators given by Bhushan et al
* (2015a)
x  ax  b (2.2)

  x*  
And a   0  , b are either real numbers or functions of the 
  w2 y 1  log   
  X*   (2.6)
known parameters of the auxiliary variables x such as the    
standard deviations S x , coefficient of variation C x , coefficient
of kurtosis  2  x  and correlation coefficient  of the Expressing T at (2.1) in terms of e ' s , we have

population,  is a suitable constant,  g ,   being constants T  w1Y 1  e0 1   ve1   w2 Y 1  e0  1   log 1  e1 
g
(2.7)
which take values  0,1, 1 for designing the different
estimators, and  w1 , w2  are suitably chosen constants to be
where v  a X a X  b
determined such that mean square error (MSE)of ' T ' is Expanding the right hand side of (2.7), neglecting terms of
minimum. It is to be mentioned that:
e ' s having higher power greater than two and subtracting Y
(i) For  w1 , w2   1, 0 , the class of estimators T can be from both sides we have

g  g  1 2 2 2 
reduced to the class of estimators due to Khoshnevisan et al.  
(2007)  w1 1  e0   vg  e1  e0 e1    v e1 
  
 
2
g T Y  Y  
(2.8)
 *

  w 1  e    e  e e     e 2   1
 
2
X 
TK  y  *  (2.3)
  x  1    X 
*  2  0 1 0 1
2 
1

 
Taking expectation on both sides of (2.8), we get the bias of
(ii) For  w1 , w2    w1 , 0 , the class of estimators T can be
the class of estimators T to the first order of approximation as
transformed to the class of estimators due to Koyuncu and
    vg  2  
 Cx  v  g  1  2k   
Kalidar (2009a)
 w1 1  
 2  
Bias(T )  Y 
g
 *
 (2.9)
K  w1 y  X       2  
 Cx  2k      1
(2.4)
  x  1    X 
* *   w2 1  
     2   

Squaring both sides of (2.8) we have

 
 
 
1  w1 1  2e0  e0  2 vge1  4 vge0 e1   v g  2 g  1 e1  
2 2 2 2 2

 2 
  w2 1  2e0  e02  2 e1   v 2    1 e12  4  ve0 e1  
 
2    2 e12 
T  Y   g  g  1    2 g 
2
 Y  2w1 w2 1  2e0  e0   e1   vge1  2  ve0 e1  2 vge0 e1 
2 2
(2.10)
  2 
  g  g  1 2 2 2  
 2w1 1  e0   vg  e1  e0 e1    v e1  
  2  
 
   2
2

 2w2 1  e0    e1  e0 e1    2 e1  
   

613
 International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

Taking expectation of both sides of (2.10), we get the MSE (2009a)),  (Bhushan et al 2015) and  (Bhushan et al
of the class of estimators ' T ' to the first order of 2015a)respectively to the first order of approximation and
approximation as will be given as:

MSE T   Y 1  w12 A  w22C  2w1w2 D  2w1B  2w2 E  (2.11)

2
MSE TK   Y 1  A  2B 
2

where
MSE TK   Y C   2 v 2 g 2Cx2  2 vgkCx2 
2 2
y (2.15)

    
A  1   C y2   2 v 2  2 g 2  g   4 vgk Cx2 
 MSE K   Y 1  w12 A  2w1 B 
2
(2.16)

  v g  g  1
  
 MSE    Y 1  C  2E 
2 2 2

B  1      vgk  Cx2 
 
 2  

MSE    Y C   2 v 2Cx2  2 vkCx2 
2 2
y

 
(2.17)
C  1   C y2    2    1  4 vk  Cx2 
 
MSE    Y 1  w22C  2w2 E 
2
(2.18)
  2  2  
 
D  1   C y   2 k     g    g  g  1  2    2 g   Cx2 
 The MSE TK  , MSE K  , MSE   and MSE   are
  2  
minimised for the optimum values

 v    ( opt ) 
k
E  1     2k  v  Cx2  vg
  2  

The MSE of the class of estimators T at (2.11) is minimised k

 ( opt ) 
for v

 BC  DE  w1(* opt ) 
B
w1   w1( opt ) (say) (2.12)
 AC  D 2 A

E
 AE  BD  w*2( opt )  respectively
and w2   w2( opt ) (say) (2.13) C
 AC  D  2

Thus, the MSEmin TK  and MSEmin K  are respectively

Substituting (2.12) and (2.13) in (2.11) we get the minimum given as
MSE of the class of estimators T as
MSEmin TK   MSEmin    Y C y2 1   2 
2

2 
MSEmin T   Y 1 
 B2C  2BDE  AE 2   (2.14)
(2.19)
  AC  D2   2 B2 
MSEmin  K   Y 1  
 A
(2.20)
Theorem 2.1. To the first order of approximation 2 E2 
MSEmin    Y 1  
2 
MSEmin T   Y 1 
 B2C  2BDE  AE 2    C 
(2.21)
  AC  D2   The MSEmin TK  , MSEmin   given by (2.19) is same as that
With equality holding if of the approximate variance/MSE of the known regression
estimator,
w1  w1( opt )

w2  w2( opt )
ylr  y   X  x  
Putting  w1 , w2   1,0 ,  w1 ,0  ,  0,1 ,  0, w2  in equation
(2.10) we will get the MSEs of the class of estimators TK
(Khoshnevisan et al. (2007)),  K (Koyuncu and Kalidar

614
 International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

3. PARTICULAR CASE  w1  w2  1 4. EFFICIENCY COMPARISON

For comparison of the efficiency of the proposed class of
For  w1  w2  1 , the proposed class of estimators estimators T with the known unbiased mean per unit
transforms to
estimator y ,usual ratio estimator y R , usual product

 
g
 X
*
  x*  y P , we have used their respective MSEs to the
T *  w1 y  *    1 
1  w y 1  log    estimator
  x  1    X *   X*  
      first order of approximation,

   
2
Var y  MSE y  Y C y2
(3.1) (4.1)

Theorem 3.1
 
2
MSE y R  Y Cy2  Cx2  2 yx C y Cx 
(4.2)
The bias and MSE of the proposed class of estimators, to the

 
first order of approximation is given by 2
MSE y P  Y C y2  Cx2  2 yx Cy Cx 
 
(4.3)
Bias T * 
  g  g  1 2 2 2
 w1   v Cx   vgkCx2 
 v 2 Cx2 
  kCx2  (3.2) 
Var y   MSEmin TK   MSEmin     MSE y lr 
   
  2 2  2 (4.4)
Y   Y C y2  2yx  0

   v 2 2
C  
x
  kCx2 
  2  
   
MSE y R  MSE y lr  Y Cx2 1  k   0
2 2

(4.5)
1  C  2 E  w12  A  C  2 D 
 
2
MSE T *
Y   (3.3)
 2w1  C  D  B  E  
    
MSE y P  MSE y lr  Y Cx2 1  k   0
2 2

(4.6)
Corollary 3.2

The resultant MSE T *  is minimised for the optimum value

MSE TK    MSEmin TK   MSE y lr 
    (4.7)
 Y Cx2  g  k   0
2 2

 C  D  B  E  **
w1   w1( opt ) (3.4)
 A  C  2D  MSE      MSEmin     MSE y lr 
    (4.8)
MSE T *  is obtained as  Y Cx2    k   0
2 2
And the minimum

 A  B
2

MSE TK   MSEmin K   Y

2
0
 C  D  B  E  
2
A
MSEmin T   Y 1  C  2E 
2 (4.9)
*
 (3.5)
  A  C  2D  
C  E 
2

MSE    MSEmin    Y
2
0
 
 Y C y2 1   2   MSE y lr E
2
(3.6) (4.10)

 AE  BD 
2
Therefore, the following theorem is established.
MSEmin  K   MSEmin T   Y
2
0
A  AC  D 2 
(4.11)
Theorem 3.3. To the first order of approximation
 BC  DE 
2

MSEmin    MSEmin T   Y
2
0
2  C  D  B  E   C  AC  D 2 
2

MSEmin T *   Y 1  C  2 E   (4.12)
  A  C  2D  
with equality holding if
 
MSEmin T *  MSEmin T  

C  B  A  D  D  B   E  A  D  
2
(4.13)
w1  w Y 
2
0
**


AC  D 2  A  C  2 D 
1( opt )

615
 International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

From (4.1)-(4.13), we get the following inequalities: MSEmin T   MSEmin K   MSE TK  (4.16)

 
MSEmin T   MSEmin T *   MSE y R  Var y  (4.14) MSEmin T   MSEmin    MSE   (4.17)
1
k
2

 
MSEmin T   MSEmin T *   MSE y P  Var y  (4.15)
1
k
2

Table 1: Some members of the class of estimators 'T '

Ratio-type logarithmic estimator  g ,     0, 1 a b
 X 
Trl1  w1 y  w2 y 1  log    1 0
  x  

  X  Cx  
Trl 2  w1 y  w2 y 1  log    1 Cx
  x  Cx  

    x  . X  Cx  
Trl 3  w1 y  w2 y 1  log  2  2  x  Cx
    x  .x  C  
 2 x 

  C .X  2  x  
Trl 4  w1 y  w2 y 1  log  x  Cx 2  x 
  C .x    x   
 x 2 

  X  Sx 
Trl 5  w1 y  w2 y 1  log    1 Sx
  x  S x  

    x  X  S x 
Trl 6  w1 y  w2 y 1  log  1  1  x  Sx
   x  x  S 
  1 x 

    x  X  S x 
Trl 7  w1 y  w2 y 1  log  2  2  x  Sx
    x  x  S 
 2 x 

  X   
Trl 8  w1 y  w2 y 1  log    1 
  x    

  X  2  x  
Trl 9  w1 y  w2 y 1  log   1 2  x 
  x    x  
 2 

  C X   
Trl10  w1 y  w2 y 1  log  x   Cx 
  Cx x    

   X  Cx  
Trl11  w1 y  w2 y 1  log     Cx
   x  Cx  

    x  X   
Trl12  w1 y  w2 y 1  log  2 
   x  x    2  x  
  2 

   X  2  x  
Trl13  w1 y  w2 y 1  log  
  x    x  
 2  x 
  2 

616
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

Table 2: Some members of the class of estimators 'T '

Product-type logarithmic estimator  g ,     0,1 a b

  x 
Tpl1  w1 y  w2 y 1  log   
 X   1 0

  x  Cx  
Tpl 2  w1 y  w2 y 1  log   
  X  Cx   1 Cx

    x  .x  Cx  
Tpl 3  w1 y  w2 y 1  log  2 
    x .X  C  2  x 
 2 x  Cx

  C .x   2  x   
Tpl 4  w1 y  w2 y 1  log  x 
  C .X    x   2  x 
 x 2  Cx

  x  S x 
Tpl 5  w1 y  w2 y 1  log   
  X  S x   1 Sx

    x  .x  S x  
Tpl 6  w1 y  w2 y 1  log  1 
    x .X  S  1  x 
 1 x  Sx

    x  .x  S x  
Tpl 7  w1 y  w2 y 1  log  2 
   x . X  S 
  2 x  2  x  Sx

  x   
Tpl 8  w1 y  w2 y 1  log   
  X     1 

  x  2  x  
Tpl 9  w1 y  w2 y 1  log  
  X    x   2  x 
 2  1

  C .x    
Tpl10  w1 y  w2 y 1  log  x  
  Cx . X     Cx 

   x  Cx  
Tpl11  w1 y  w2 y 1  log   
   X  Cx    Cx

    x  .x    
Tpl12  w1 y  w2 y 1  log  2 
    x . X   
 2  2  x  

   x  2  x  
Tpl13  w1 y  w2 y 1  log  

  X    x  
 2   2  x 

617
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

Table 3: Some members of the class of estimators 'T '

Ratio-ratio type logarithmic estimator  , g ,    1,1, 1 a b

X   X 
Trrl1  w1 y    w2 y 1  log   
 x    x   1 0

 X  Cx    X  Cx  
Trrl 2  w1 y    w2 y 1  log   
 x  Cx    x  Cx   1 Cx

   x  X  Cx      x  . X  Cx  
Trrl 3  w1 y  2   w2 y 1  log  2 
   x x  C     x  .x  C   2  x 
 2 x    2 x 
Cx

 C X  2  x     C .X  2  x  
Trrl 4  w1 y  x   w2 y 1  log  x 
 C x    x    C .x    x    Cx 2  x 
 x 2   x 2 

 X  Sx    X  Sx 
Trrl 5  w1 y    w2 y 1  log 
  
 x  Sx    x  S x   1 Sx

   x  X  Sx      x .X  S x 
Trrl 6  w1 y  1   w2 y 1  log  1 
   x x  S      x  .x  S   1  x  Sx
 1 x   1 x 

   x  X  Sx      x .X  S x 
Trrl 7  w1 y  2   w2 y 1  log  2 
   x x  S      x  .x  S   2  x  Sx
 2 x   2 x 

X    X   
Trrl 8  w1 y    w2 y 1  log   
 x    x     1 

 X  2  x     X  2  x  
Trrl 9  w1 y    w2 y 1  log  
 x    x    x    x   1 2  x 
 2   2 

C X      C .X   
Trrl10  w1 y  x   w2 y 1  log  x  
 Cx x      Cx .x     Cx 

  X  Cx     X  Cx  
Trrl11  w1 y    w2 y 1  log   
  x  Cx     x  Cx    Cx

   x X        x .X   
Trrl12  w1 y  2   w2 y 1  log  2 
   x x   
 2      x  .x    
 2  2  x  

  X  2  x      X  2  x  
Trrl13  w1 y    w2 y 1  log  
  x    x 
 2  
  x    x  
 2 
 2  x 

618
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

Table 4: Some members of the class of estimators 'T '

ratio-product type logarithmic estimators  , g ,    1,1,1 a b

X    x 
Trpl1  w1 y    w2 y 1  log   
 x   X   1 0

 X  Cx    x  Cx  
Trpl 2  w1 y    w2 y 1  log   
 x  Cx    X  Cx   1 Cx

   x  X  Cx      x  .x  C x  
Trpl 3  w1 y  2   w2 y 1  log  2 
   x x  C     x .X  C  2  x 
 2 x    2 x 
Cx

 C X  2  x     C .x   2  x   
Trpl 4  w1 y  x   w2 y 1  log  x 
 C x    x    C .X    x   Cx 2  x 
 x 2   x 2 

 X  Sx    x  S x 
Trpl 5  w1 y    w2 y 1  log 
  
 x  Sx    X  S x   1 Sx

   x  X  Sx      x  .x  S x  
Trpl 6  w1 y  1   w2 y 1  log  1 
   x x  S      x .X  S  1  x  Sx
 1 x   1 x 

   x  X  Sx      x  .x  S x  
Trpl 7  w1 y  2   w2 y 1  log  2 
   x x  S      x .X  S  2  x  Sx
 2 x   2 x 

X    x   
Trpl 8  w1 y    w2 y 1  log   
 x    X     1 

 X  2  x     x  2  x  
Trpl 9  w1 y    w2 y 1  log  
 x    x    X    x   1 2  x 
 2   2 

C X      C .x    
Trpl10  w1 y  x   w2 y 1  log  x  
 Cx x      Cx . X     Cx 

  X  Cx     x  Cx  
Trpl11  w1 y    w2 y 1  log   
  x  Cx     X  Cx    Cx

   x X        x  .x    
Trpl12  w1 y  2   w2 y 1  log  2 
   x x   
 2      x .X   
 2  2  x  

  X  2  x      x  2  x  
Trpl13  w1 y    w2 y 1  log  
  x    x 
 2  
  X    x  
 2 
 2  x 

619
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

Table 5: Some members of the class of estimators 'T '

Product-product type logarithmic estimators  , g ,    1, 1,1 a b

 x    x 
Tppl1  w1 y    w2 y 1  log   
X   X   1 0

 x  Cx    x  Cx  
Tppl 2  w1 y    w2 y 1  log   
 X  Cx    X  Cx   1 Cx

   x  x  Cx      x  .x  C x  
Tppl 3  w1 y  2   w2 y 1  log  2 
   x X  C     x .X  C 
 2 x    2 x  2  x  Cx

 C x  2  x     C x  2  x  
Tppl 4  w1 y  x   w2 y 1  log  x 
 C X    x   C X    x  
 x 2    x 2  Cx 2  x 

 x  Sx    x  S x 
Tppl 5  w1 y    w2 y 1  log   
 X  Sx    X  S x   1 Sx

   x  x  Sx      x  .x  S x  
Tppl 6  w1 y  1   w2 y 1  log  1 
   x X  S     x .X  S 
 1 x    1 x  1  x  Sx

   x  x  Sx      x  .x  S x  
Tppl 7  w1 y  2   w2 y 1  log  2 
   x X  S     x .X  S 
 2 x    2 x  2  x  Sx

 x    x   
Tppl 8  w1 y    w2 y 1  log   
X    X     1 

 x  2  x     x  2  x  
Tppl 9  w1 y    w2 y 1  log  
 X    x   X    x  
 2    2  1 2  x 

 C x    C .x    
Tppl10  w1 y  x   w2 y 1  log  x  
 x
C X      Cx . X     Cx 

  x  Cx     x  Cx  
Tppl11  w1 y    w2 y 1  log 
  
  X  Cx     X  Cx    Cx

   x x        x  .x    
Tppl12  w1 y  2   w2 y 1  log  2 
   x X   
 2      x .X   
 2  2  x  

  x  2  x      x  2  x  
Tppl13  w1 y    w2 y 1  log  
  X    x 
 2     X    x  
 2   2  x 

620
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

Table 6: Some members of the class of estimators 'T '

Product-ratio type logarithmic estimators  , g ,    1, 1,1 a b

 x   X 
Tprl1  w1 y    w2 y 1  log   
X    x   1 0

 x  Cx    X  Cx  
Tprl 2  w1 y    w2 y 1  log   
 X  Cx    x  Cx   1 Cx

   x  x  Cx      x  . X  Cx  
Tprl 3  w1 y  2   w2 y 1  log  2 
   x X  C     x  .x  C  
 2 x    2 x  2  x  Cx

 C x  2  x     C .X  2  x  
Tprl 4  w1 y  x   w2 y 1  log  x 
 C X    x   C .x    x   
 x 2    x 2  Cx 2  x 

 x  Sx    X  S x 
Tprl 5  w1 y    w2 y 1  log  
 X  Sx    x  S x  1 Sx

   x  x  Sx      x .X  S x 
Tprl 6  w1 y  1   w2 y 1  log  1 
   x X  S     x  .x  S  
 1 x    1 x  1  x  Sx

   x  x  Sx      x .X  S x 
Tprl 7  w1 y  2   w2 y 1  log  2 
   x X  S     x  .x  S  
 2 x    2 x  2  x  Sx

 x    X   
Tprl 8  w1 y    w2 y 1  log   
X    x     1 

 x  2  x     X  2  x  
Tprl 9  w1 y    w2 y 1  log  
 X    x   x    x  
 2    2  1 2  x 

 C x    C .X   
Tprl10  w1 y  x   w2 y 1  log  x  
 x
C X      Cx .x     Cx 

  x  Cx     X  Cx  
Tprl11  w1 y    w2 y 1  log 
  
  X  Cx     x  Cx    Cx

   x x        x .X   
Tprl12  w1 y  2   w2 y 1  log  2 
   x X   
 2      x  .x    
 2  2  x  

  x  2  x      X  2  x  
Tprl13  w1 y    w2 y 1  log  
  X    x 
 2     x    x  
 2   2  x 

621
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

Table 7: Some members of the class of estimators TK

Ratio type estimators Product type estimators a b

 , g   1,1  , g   1, 1

X   x 
TKr1  w1 y   TKp1  w1 y  
 x  X  1 0

 X  Cx   x  Cx 
TKr 2  w1 y   TKp 2  w1 y  
 x  Cx   X  Cx  1 Cx

   x  X  Cx     x  x  Cx 
TKr 3  w1 y  2  TKp 3  w1 y  2 
   x x  C     x X  C  2  x 
 2 x   2 x  Cx

 C X  2  x    C x  2  x  
TKr 4  w1 y  x  TKp 4  w1 y  x 
 C x    x   C X    x  2  x 
 x 2   x 2  Cx

 X  Sx   x  Sx 
TKr 5  w1 y   TKp 5  w1 y  
 x  Sx   X  Sx  1 Sx

   x  X  Sx     x  x  Sx 
TKr 6  w1 y  1  TKp 6  w1 y  1 
   x x  S     x X  S  1  x 
 1 x   1 x  Sx

   x  X  Sx     x  x  Sx 
TKr 7  w1 y  2  TKp 7  w1 y  2 
   x x  S     x X  S  2  x 
 2 x   2 x  Sx

X   x 
TKr 8  w1 y   TKp8  w1 y  
 x  X  1 

 X  2  x    x  2  x  
TKr 9  w1 y   TKp 9  w1 y  
 x    x   X    x  2  x 
 2   2  1

C X     C x 
TKr10  w1 y  x  TKp10  w1 y  x 
 Cx x     Cx X    Cx 

  X  Cx    x  Cx 
TKr11  w1 y   TKp11  w1 y  
  x  Cx    X  Cx   Cx

   x X       x x   
TKr12  w1 y  2  TKp12  w1 y  2 
   x x   
 2 
   x X   
 2  2  x  

  X  2  x     x  2  x  
TKr13  w1 y   TKp13  w1 y  
  x    x 
 2 
  X    x 
 2   2  x 

622
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

Table 8: Some members of the class of estimators  K

Ratio type estimators Product type estimators a b

 , g   1,1  , g   1, 1

X  x 
 K r1  w1 y    K p1  w1 y  
 x X  1 0

 X  Cx   x  Cx 
 K r 2  w1 y    K p 2  w1 y  
 x  Cx   X  Cx  1 Cx

  2  x  X  Cx    2  x  x  Cx 
 K r 3  w1 y    K p 3  w1 y  
   x x  C     x X  C  2  x 
 2 x   2 x  Cx

 Cx X   2  x    Cx x   2  x  
 K r 4  w1 y    K p 4  w1 y  
 C x    x   C X    x  2  x 
 x 2   x 2  Cx

 X  Sx   x  Sx 
 K r 5  w1 y    K p 5  w1 y  
 x  Sx   X  Sx  1 Sx

 1  x  X  S x   1  x  x  S x 
 K r 6  w1 y    K p 6  w1 y  
   x x  S     x X  S  1  x 
 1 x   1 x  Sx

 2  x  X  S x   2  x  x  Sx 
 K r 7  w1 y    K p 7  w1 y  
   x x  S     x X  S  2  x 
 2 x   2 x  Sx

X   x 
 K r 8  w1 y    K p8  w1 y  
 x  X  1 

 X  2  x    x  2  x  
 K r 9  w1 y    K p 9  w1 y  
 x    x   X    x  2  x 
 2   2  1

 Cx X     Cx x   
 K r10  w1 y    K p10  w1 y  
 Cx x     Cx X    Cx 

  X  Cx    x  Cx 
 K r11  w1 y    K p11  w1 y  
  x  Cx    X  Cx   Cx

 2  x  X     2  x  x   
 K r12  w1 y    K p12  w1 y  
   x x   
 2 
   x X   
 2  2  x  

  X  2  x     x  2  x  
 K r13  w1 y    K p13  w1 y  
  x    x 
 2 
  X    x 
 2   2  x 

623
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

Table 9: Some members of the class of estimators 

Ratio type estimators Product type estimators a b

   1    1

  X    x 
 r1  y 1  log     p1  y 1  log   
  x    X   1 0

  X  Cx     x  Cx  
 r 2  1  log     p 2  y 1  log   
  x  Cx     X  Cx   1 Cx

   2  x  . X  Cx      2  x  .x  C x  
r 3  1  log    p 3  y 1  log  
    x  .x  C       x .X  C  2  x  Cx
 2 x   2 x 

  Cx . X   2  x      C x .x   2  x   
r 4  1  log    p 4  y 1  log  
  C .x    x      C .X    x   Cx 2  x 
 x 2   x 2 

  X  Sx    x  Sx 
 r 5  1  log     p 5  y 1  log   
  x  S x     X  S x   1 Sx

  1  x  . X  S x     1  x  .x  S x  
r 6  1  log    p 6  y 1  log  
   x  .x  S       x .X  S  1  x 
  1 x   1 x  Sx

  2  x .X  S x     2  x  .x  S x  
r 7  1  log    p 7  y 1  log  
    x  .x  S       x .X  S  2  x  Sx
 2 x   2 x 

  X      x   
 r 8  1  log     p8  y 1  log   
  x       X     1 

  X  2  x     x  2  x  
 r 9  1  log    p 9  y 1  log  
 x    x    X    x   2  x 
  2    2  1

  Cx . X       Cx x    
 r10  1  log     p10  y 1  log   
  Cx .x       Cx X     Cx 

   X  Cx      x  Cx  
 r11  1  log     p11  y 1  log   
   x  Cx      X  Cx    Cx

  2  x .X       2  x  .x    
r12  1  log    p12  y 1  log  
    x  .x    
 2      x .X   
 2  2  x  

   X  2  x      x  2  x  
r13  1  log    p13  y 1  log  
   x    x  
 2     X    x  
 2   2  x 

624
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

Table 10: Some members of the class of estimators 

Ratio type estimators Product type estimators a b

   1    1

  X    x 
 r1  w2 y 1  log     p1  w2 y 1  log   
  x    X   1 0

  X  Cx     x  Cx  
 r 2  w2 y 1  log     p 2  w2 y 1  log   
  x  Cx     X  Cx   1 Cx

   2  x  . X  Cx      2  x  .x  C x  
 r 3  w2 y 1  log    p 3  w2 y 1  log  
    x  .x  C       x .X  C  2  x  Cx
 2 x   2 x 

  Cx . X   2  x      Cx .x   2  x   
 r 4  w2 y 1  log    p 4  w2 y 1  log  
  C .x    x      C .X    x   Cx 2  x 
 x 2   x 2 

  X  Sx    x  S x 
 r 5  w2 y 1  log     p 5  w2 y 1  log   
  x  S x     X  S x   1 Sx

  1  x  . X  S x     1  x  .x  S x  
 r 6  w2 y 1  log    p 6  w2 y 1  log  
   x  .x  S       x .X  S  1  x 
  1 x   1 x  Sx

  2  x .X  Sx     2  x  .x  S x  
 r 7  w2 y 1  log    p 7  w2 y 1  log  
    x  .x  S       x .X  S  2  x  Sx
 2 x   2 x 

  X      x   
 r 8  w2 y 1  log     p8  w2 y 1  log   
  x       X     1 

  X  2  x     x  2  x  
 r 9  w2 y 1  log    p 9  w2 y 1  log  
 x    x    X    x   2  x 
  2    2  1

  Cx . X       Cx x    
 r10  w2 y 1  log     p10  w2 y 1  log   
  Cx .x       Cx X     Cx 

   X  Cx      x  Cx  
 r11  w2 y 1  log     p11  w2 y 1  log   
   x  Cx      X  Cx    Cx

  2  x . X       2  x  .x    
 r12  w2 y 1  log    p12  w2 y 1  log  
    x  .x    
 2      x .X   
 2  2  x  

   X  2  x      x  2  x  
 r13  w2 y 1  log    p13  w2 y 1  log  
   x    x  
 2     X    x  
 2   2  x 

625
 International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

5. EMPIRICAL STUDY REFERENCES

To examine the merits of the suggested class of estimators [1] Bhushan. S., Gupta. R and Pandey. S.K. (2015): Some
' T ' we considered a natural population data set. The log-type classes of estimators using auxiliary
information (Submitted).
description of the population is given below.
[2] Bhushan. S., Gupta. R and Pandey. S.K. (2015a):
Population 1: (Singh, D and Chaudhary, F.S). The data Improved Searls type logarithmic estimators using
concerns the no. of bearing lime trees and the area reported auxiliary information (Submitted).
under lime, in each of the 22 villages growing lime in one of
[3] Kalidar. C, Cingi. H. (2006). An improvement in
the tehsils of Nellore district. Pg.no.141
estimating the population mean by using the correlation
y : area reported under lime coefficient. Hacett. J. Math. Statist. 35(1): 103-109.
[4] Khoshnevisan.M., Singh R., Chauhan. P., Sawan. N.,
x : no. of bearing lime trees Smarandache. F. (2007): A general family of estimators
for estimating population mean using known value of
N  22 ; n  15 ; Y  22.62091 ; X  1467.545 some population parameter(s). Far East J. Theor.
Statist. 22: 181-191.
S y  33.04697 ; S x  2562.145 ;   0.902147
[5] Koyuncu. N., Kalidar.C. (2009a): Efficient estimators
for the population mean. Hacett. J. Math. Statist. 38(2):
217-225.
Population2: ( Koyuncu and Kalidar, 2009a). The data
concerns primary and secondary school for 923 districts of [6] Pandey. B. N., Dubey. V.(1988): Modified product
Turkey in 2007. The variables are defined as follows: estimator using coefficient of variation of auxiliary
variate. Assam Statist. Rev. 2(2): 64-66
y : number of teachers in both primary and secondary school [7] Searls.D.T. (1964): The utilization of known
coefficient of variation in the estimation procedure. J.
x : number of students in both primary and secondary school.
Amer. Statist Assoc.59: 1225-1226.
N  923 ; n  180 ; Y  436.4345 ; X  11440.4984 [8] Singh, G. N. (2003): On the improvement of product
method of estimation in sample surveys. J.Ind. Soc.
S y  749.9395 ; Sx  2131.1315 ;   0.9543 Agri. Statis. 56(3): 265-267.
[9] Singh. H. P., Tailor. R. (2003): Use of known
correlation coefficient in estimating the finite
Population 1 Population 2 population mean. Statist. Trans. 6: 555-560.

ESTIMATORS MSE PRE MSE PRE [10] Singh. H. P., Agnihotri. N. (2008): A general procedure
of estimating population mean using auxiliary
y 23.16581 100 2515.168 100 information in sample surveys. Statist. Trans. 9(1): 71-
87
yR 6.299483 367.74 267.6515 939.7177 [11] Singh. H. P., Tailor. R., Kakran. M.S. (2004): An
estimator of population mean using power
yP 106.2017 21.813 10685.4 23.53837 transformation. J.Ind. Soc. Agri. Statis. 58(2): 223-230.
[12] Singh. H. P., Tailor. R., Singh. S., Kim. J. M. (2008): A
TK 4.311873 537.26 224.6335 1119.677 modified estimator of population mean using power
transformation. Statist. Pap. 49: 37-58.
K 4.275842 541.78 224.3689 1120.997
[13] Upadhyaya. L. N., Singh. H. P. (1999): Use of
 4.311873 537.26 224.6335 1119.677 transformed auxiliary variable in estimating the finite
population mean. Biometrical J. 41: 27-36.
 4.303488 538.3 221.5432 1135.295
[14] Bhushan.S., Misra.P.K and Yadav.S.K. (2017a): A
T* 4.311873 537.26 224.6335 1119.677 Generalized Class of Unbiased Estimators for
Population Mean Using Auxiliary Information on an
T 4.311873 537.26 224.6335 1119.677 Attribute and an Auxiliary Variable, International
Journal of Computational and Applied Mathematics,
Note: The above calculations are made on the basis of the Volume 12, Number 1, 11-28.
optimum values of the characterising parameters  ,   and
[15] Bhushan.S., Misra.P.K and Yadav.S.K. (2017b): On
 g  1 . Others calculations can be made on the similar lines. Unbiased Class of Ratio Estimator for Population Mean
Using Auxiliary Information on an Attribute and a

626
 International Journal of Applied Engineering Research ISSN 0973-4562 Volume 15, Number 6 (2020) pp. 612-627
© Research India Publications. http://www.ripublication.com

Variable, International Journal of Statistics and

Systems, Volume 12, Number 1, 25-32.
[16] Bhushan.S., Misra.P.K and Yadav.S.K. (2017c): An
Unbiased Estimator For Population Mean Using An
Attribute And An Auxiliary Variable, Advances in
Dynamical Systems and Applications, Volume 12,
Number 1, 29-39
[17] Bhushan.S., Misra.P.K . (2017d): A Family of
Unbiased Estimators of Population Mean Using an
Auxiliary Variable, Advances in Computational
Sciences and Technology ,Volume 10, Number 1, 129-
137.
[18] Bhushan, S., Gupta.R. (2020): An improved log-type
family of estimators using Attribute, Journal of
Statistics & Management Systems
[19] Bhushan, S., Gupta.R. (2019a): A class of log-type
estimators for population mean using auxiliary
information on an attribute and a variable using double
sampling technique. International Journal of
Computational and Applied Mathematics, vol 14, no.1,
pp 1-12
[20] Bhushan, S., Gupta.R. (2019b): Searls’ Ratio Product
Type Estimators. International Journal of Statistics and
Systems, Volume 14, Number 1, pp. 29-37
[21] Bhushan, S., Gupta.R. (2019c): Some Log-Type
Classes of Estimators Using Auxiliary Attribute,
Advances in Computational Sciences and Technology
,Volume 12, Number 2, pp. 99-108
[22] Bhushan, S., Gupta.R. (2019d): Some New Log Type
Class of Double Sampling Estimators, International
Journal of Applied Agricultural Research, Volume 14,
Number 1, pp. 31-40

627