Chapter 7

Varying Probability Sampling

The simple random sampling scheme provides a random sample where every unit in the population has
equal probability of selection. Under certain circumstances, more efficient estimators are obtained by
assigning unequal probabilities of selection to the units in the population. This type of sampling is
known as varying probability sampling scheme.

If Y is the variable under study and X is an auxiliary variable related to Y, then in the most commonly
used varying probability scheme, the units are selected with probability proportional to the value of X,
called as size. This is termed as probability proportional to a given measure of size (pps) sampling. If
the sampling units vary considerably in size, then SRS does not takes into account the possible
importance of the larger units in the population. A large unit, i.e., a unit with large value of Y contributes
more to the population total than the units with smaller values, so it is natural to expect that a selection
scheme which assigns more probability of inclusion in a sample to the larger units than to the smaller
units would provide more efficient estimators than the estimators which provide equal probability to all
the units. This is accomplished through pps sampling.

Note that the “size” considered is the value of auxiliary variable X and not the value of study variable Y.
For example in an agriculture survey, the yield depends on the area under cultivation. So bigger areas are
likely to have larger population and they will contribute more towards the population total, so the value
of the area can be considered as the size of auxiliary variable. Also, the cultivated area for a previous
period can also be taken as the size while estimating the yield of crop. Similarly, in an industrial survey,
the number of workers in a factory can be considered as the measure of size when studying the industrial
output from the respective factory.

Difference between the methods of SRS and varying probability scheme:

In SRS, the probability of drawing a specified unit at any given draw is the same. In varying probability
scheme, the probability of drawing a specified unit differs from draw to draw.
It appears in pps sampling that such procedure would give biased estimators as the larger units are over-
represented and the smaller units are under-represented in the sample. This will happen in case of
sample mean as an estimator of population mean where all the units are given equal weight. Instead of
giving equal weights to all the units, if the sample observations are suitably weighted at the estimation

1
stage by taking the probabilities of selection into account, then it is possible to obtain unbiased
estimators.

In pps sampling, there are two possibilities to draw the sample, i.e., with replacement and without
replacement.

Selection of units with replacement:

The probability of selection of a unit will not change and the probability of selecting a specified unit is
same at any stage. There is no redistribution of the probabilities after a draw.

Selection of units without replacement:

The probability of selection of a unit will change at any stage and the probabilities are redistributed after
each draw.

PPS without replacement (WOR) is more complex than PPS with replacement (WR) . We consider both
the cases separately.

PPS sampling with replacement (WR):

First we discuss the two methods to draw a sample with PPS and WR.

1. Cumulative total method:

The procedure of selection a simple random sample of size n consists of
- associating the natural numbers from 1 to N units in the population and
- then selecting those n units whose serial numbers correspond to a set of n numbers where each
number is less than or equal to N which is drawn from a random number table.

In selection of a sample with varying probabilities, the procedure is to associate with each unit a set of
consecutive natural numbers, the size of the set being proportional to the desired probability.

If X 1 , X 2 ,..., X N are the positive integers proportional to the probabilities assigned to the N units in the

population, then a possible way to associate the cumulative totals of the units. Then the units are selected
based on the values of cumulative totals. This is illustrated in the following table:

2
Units Size Cumulative
1 X1 T1  X 1
 If Ti 1  R  Ti , then
2 X2 T2  X 1  X 2 ith unit is selected
  Select a random with probability
 number R
i 1
Xi
between 1 and TN , i = 1,2,…, N .
i 1 X i 1 Ti 1   X j TN
j 1 by using random
number table.
i  Repeat the procedure
i Xi Ti   X j n times to get a
 
j 1
sample of size n.

N
XN   X j
N
N TN   X j
j 1 j 1

In this case, the probability of selection of ith unit is

Ti  Ti 1 X i
Pi  
TN TN
 Pi  X i .

Note that TN is the population total which remains constant.

Drawback : This procedure involves writing down the successive cumulative totals. This is time
consuming and tedious if the number of units in the population is large.

This problem is overcome in the Lahiri’s method.

Lahiri’s method:
Let M  Max X i , i.e., maximum of the sizes of N units in the population or some convenient
i 1,2,..., N

number greater than M .

The sampling procedure has following steps:
1. Select a pair of random number (i, j) such that 1  i  N , 1  j  M .
2. If j  X i , then ith unit is selected otherwise rejected and another pair of random number is

chosen.
3. To get a sample of size n , this procedure is repeated till n units are selected.
Now we see how this method ensures that the probabilities of selection of units are varying and are
proportional to size.

3
Probability of selection of ith unit at a trial depends on two possible outcomes
– either it is selected at the first draw
– or it is selected in the subsequent draws preceded by ineffective draws. Such probability is given by
P (1  i  N ) P(1  j  M | i )
1 X
 . i  Pi * , say.
N M
1 N
 Xi 
Probability that no unit is selected at a trial 
N
 1  M 
i 1

1 NX 
 N  
N M 
X
 1  Q, say.
M
Probability that unit i is selected at a given draw (all other previous draws result in the non selection of
unit i)
 Pi *  QPi *  Q 2 Pi *  ...
Pi *

1 Q
X / NM X Xi
 i  i   Xi.
X /M NX X total

Thus the probability of selection of unit i is proportional to the size X i . So this method generates a pps

sample.

Advantage:
1. It does not require writing down all cumulative totals for each unit.
2. Sizes of all the units need not be known before hand. We need only some number greater than the
maximum size and the sizes of those units which are selected by the choice of the first set of
random numbers 1 to N for drawing sample under this scheme.

Disadvantage: It results in the wastage of time and efforts if units get rejected.
X
The probability of rejection  1  .
M
M
The expected numbers of draws required to draw one unit  .
X
This number is large if M is much larger than X .

4
Example: Consider the following data set of 10 number of workers in the factory and its output. We
illustrate the selection of units using the cumulative total method.

Factory no. Number of workers Industrial production Cumulative total of sizes

(X) (in thousands) (in metric tonns) (Y)
1 2 30 T1  2

2 5 60 T2  2  5  7

3 10 12 T3  2  5  10  17

4 4 6 T4  17  4  21

5 7 8 T5  21  7  28

6 12 13 T6  28  12  30

7 3 4 T7  30  3  33

8 14 17 T8  33  14  47

9 11 13 T9  47  11  58

10 6 8 T10  58  6  64

Selection of sample using cumulative total method:

1.First draw: - Draw a random number between 1 and 64.
- Suppose it is 23
- T4  23  T5

- Unit Y is selected and Y5  8 enters in the sample .

2. Second draw:
- Draw a random number between 1 and 64
- Suppose it is 38
- T7  38  T8

- Unit 8 is selected and Y8  17 enters in the sample

- and so on.
- This procedure is repeated till the sample of required size is obtained.

5
Selection of sample using Lahiri’s Method
In this case
M  Max X i  14
i 1,2,...,10

So we need to select a pair of random number (i, j ) such that 1  i  10, 1  j  14 .

Following table shows the sample obtained by Lahiri’s scheme:
Random no Random no Observation Selection of unit
1  i  10 1  j  14

3 7 j  7  X 3  10 trial accepted ( y3 )

6 13 j  13  X 6  12 trial rejected

4 7 j  7  X4  4 trial rejected

2 9 j  9  X2  5 trial rejected

9 2 j  2  X 9  11 trial accepted ( y9 )

and so on. Here ( y3 , y9 ) are selected into the sample.

Varying probability scheme with replacement: Estimation of population mean

Let
Yi : value of study variable for the ith unit of the population, i = 1, 2,…,N.
X i : known value of auxiliary variable (size) for the ith unit of the population.

Pi : probability of selection of ith unit in the population at any given draw and is proportional to size X i .

Consider the varying probability scheme and with replacement for a sample of size n. Let yr be the

value of rth observation on study variable in the sample and pr be its initial probability of selection.

Define
yr
zr  , r  1, 2,..., n,
Npr
then
1 n
z  zi
n i 1

6
 2
 z2 N
 Y 
is an unbiased estimator of population mean Y , variance of z is where    Pi  i  Y  and
2
z
n i 1  NPi 
sz2 1 n
an unbiased estimate of variance of z is  
n n  1 r 1
( zr  z ) 2 .

Proof:
Note that zr can take any one of the N values out of Z1 , Z 2 ,..., Z N with corresponding initial probabilities

P1 , P2 ,..., PN , respectively. So
N
E ( zr )   Z i Pi
i 1

N
Yi
 Pi
i 1 NPi

Y.
Thus
1 n
E(z )   E ( zr )
n i 1
1 n
 Y
n i 1

Y.

So z is an unbiased estimator of population mean Y .

The variance of z is
1  n 
Var ( z ) 
n2
Var   zr 
 r 1 
n
1

n2
Var ( z )
r 1
r ( zr' s are independent in WR case).

Now

Var ( zr )  E  zr  E ( zr ) 
2

2
 E  zr  Y 
N
   Z i  Y  Pi
2

i 1

2
 Y
N

   i  Y  Pi
i 1  NPi 
  z2 (say) .

7
Thus
n
1
Var ( z ) 
n2

r 1
2
z

 2
 z
.
n

sz2
To show that is an unbiased estimator of variance of z , consider
n
 n 
(n  1) E ( sz2 )  E   ( zr  z ) 2 
 r 1 
 n 
 E   zr2  nz 2 
 r 1 
 n 
   E ( zr2 )  nE ( z ) 2 
 r 1 
n
  Var ( zr )   E ( zr )   n Var ( z )   E ( z ) 
2 2

r 1
   

 2

  n   Y 
    Y
n N
2 2  z2
Y 2
 using Var ( zr )    i  Y  Pi   z2 
z n
 i 1  NPi  
r 1
 
 (n  1) z2
E ( sz2 )   z2
 sz2   z2
or E    Var ( z )
n n

1  n  yr  
2
sz2

 Var ( z )       nz 2
.
n n(n  1)  r 1  Npr  

1
Note: If Pi  , then z  y ,
N
2
 
1 1 N  Yi   y2
Var ( z )  
n N i 1  N . 1
 Y  
n

 N 
which is the same as in the case of SRSWR.

8
Estimation of population total:
An estimate of population total is

1 n y 
Yˆtot    r   N z . .
n r 1  pr 

Taking expectation, we get

1 n Y Y Y 
E (Yˆtot )    1 P1  2 P2  ...  N PN 
n r 1  P1 P2 PN 
1 n N 
    Yi 
n r 1  i 1 
1 n
  Ytot
n r 1
 Ytot .

Thus Yˆtot is an unbiased estimator of population total. Its variance is

Var (Yˆtot )  N 2Var ( z )

2
1 N 1 Y 
 N  2  i  NY  Pi
2

n i 1 N  Pi 
2
1 N Y 
   i  Ytot  Pi
n i 1  Pi 
1  N Yi 2 
   Ytot2  .
n  i 1 Pi 

An estimate of the variance

2
 (Yˆ )  N 2 sz .
Var tot
n

Varying probability scheme without replacement

In varying probability scheme without replacement, when the initial probabilities of selection are
unequal, then the probability of drawing a specified unit of the population at a given draw changes with
the draw. Generally, the sampling WOR provides a more efficient estimator than sampling WR. The
estimators for population mean and variance are more complicated. So this scheme is not commonly
used in practice, especially in large scale sample surveys with small sampling fractions.

9
Let U i : i th unit,

Pi : Probability of selection of U i at the first draw, i  1, 2,..., N

N

 P 1
i 1
i

Pi ( r ) : Probability of selecting U i at the r th draw

Pi (1)  Pi .

Consider
Pi (2)  Probability of selection of U i at 2nd draw.

Such an event can occur in the following possible ways:

U i is selected at 2nd draw when

- U1 is selected at 1st draw and U i is selected at 2nd draw

- U 2 is selected at 1st draw and U i is selected at 2nd draw

- U i -1 is selected at 1st draw and U i is selected at 2nd draw
- U i 1 is selected at 1st draw and U i is selected at 2nd draw

- U N is selected at 1st draw and U i is selected at 2nd draw

So Pi (2) can be expressed as

Pi P Pi Pi Pi
Pi (2)  P1  P2 i  ...  Pi 1  Pi 1  ...  PN
1  P1 1  P2 1  Pi 1 1  Pi 1 1  PN
N
Pi
 
j (  i ) 1
Pj
1  Pj
N
Pi P P
 
j (  i ) 1
Pj
1  Pj
 Pi i  Pi i
1  Pi 1  Pi
N
Pi P
  Pj  Pi i
j 1 1  Pj 1  Pi
N P P 
 Pi   j  i 
 j 1 1  Pj 1  Pi 
1
Pi (2)  Pi (1) for all i unless Pi  .
N

10
 y 
Pi (2) will, in general, be different for each i = 1,2,…, N . So E  i  will change with successive draws.
 pi 
y1
This makes the varying probability scheme WOR more complex. Only will provide an unbiased
Np1
yi
estimator of Y . In general, (i  1) will not provide an unbiased estimator of Y .
Npi

Ordered estimates
To overcome the difficulty of changing expectation with each draw, associate a new variate with each
draw such that its expectation is equal to the population value of the variate under study. Such
estimators take into account the order of the draw. They are called the ordered estimates. The order of
the value obtained at previous draw will affect the unbiasedness of population mean.

We consider the ordered estimators proposed by Des Raj, first for the case of two draws and then
generalize the result.

Des Raj ordered estimator

Case 1: Case of two draws:
Let y1 and y2 denote the values of units U i (1) and U i (2) drawn at the first and second draws

respectively. Note that any one out of the N units can be the first unit or second unit, so we use the
notations U i (1) and U i (2) instead of U1 and U 2 . Also note that y1 and y2 are not the values of the first two

units in the population. Further, let p1 and p2 denote the initial probabilities of selection of Ui(1) and

Ui(2), respectively.

Consider the estimators

y1
z1 
Np1

1 y2 
z2   y1  
N p2 / (1  p1 ) 

1 (1  p1 ) 
  y1  y2 
N p2 
z1  z2
z .
2
p2
Note that is the probability P (U i (2) | U i (1) ).
1  p1
11
Estimation of Population Mean:
First we show that z is an unbiased estimator of Y .
E(z )  Y .
N
Note that  P  1.
i 1
i

Consider
1  y1   y1 Y Y Y 
E ( z1 )  E   Note that can take any one of out of the N values 1 , 2 ,..., N 
N  p1   p1 P1 P2 PN 

1  Y1 Y2 YN 
  P1  P2  ...  PN 
N  P1 P2 PN 

Y

1  (1  p1 ) 
E ( z2 )  E  y1  y2 
N  p2 

1    (1  P1 )  
  E ( y1 )  E1  E2  y2 U i (1)   (Using E (Y )  E X [ EY (Y | X )].
N    p2  

where E2 is the conditional expectation after fixing the unit U i (1) selected in the first draw.

y2 Y
Since can take any one of the (N – 1) values (except the value selected in the first draw) j with
p2 Pj

Pj
probability , so
1  P1

 (1  P1 )  y  Y P 
E2  y2 U i (1)   (1  P1 ) E2  2 U i (1)   (1  P1 )  j . j  .
 p2   p2   Pj 1  P1 
j 

where the summation is taken over all the values of Y except the value y1 which is selected at the first
draw. So
 (1  P1 ) 
E2  y2 U i (1)    Y j  Ytot  y1. .
 p2  j

Substituting it is E ( z2 ), we have

1
E ( z2 )   E ( y1 )  E1 (Ytot  y1 )
N
1
  E ( y1 )  E (Ytot  y1 )
N
1 Y
 E (Ytot )  tot  Y .
N N
12
 E ( z1 )  E ( z2 )
Thus E ( z ) 
2
Y Y

2
Y.

Variance:
The variance of z for the case of two draws is given as

 1 
2 2
 1 N N
Y  1 N
Y 
Var ( z )  1   Pi 2   2  Pi  i  Ytot    Pi  i  Ytot 
2

 2 i 1   2 N  4 N
2
i 1  Pi  i 1  Pi 

Proof: Before starting the proof, we note the following property

N N N 
 a b
i j   ai   b j  bi 
i  j 1 i 1  j 1 
which is used in the proof.

The variance of z is

Var ( z )  E ( z 2 )   E ( z )
2

2
 1  y1 y2 (1  p1 ) 
 E   y1    Y
2

 2 N  1
p p2 
2
1  y (1  p1 ) y2 (1  p1 ) 
 2
E 1   Y
2

4N  p1 p2 
 
nature of nature of
variable variable
depends depends
only on upon1st and
1st draw 2nd draw

1  N  Yi (1  Pi ) Y j (1  Pi )  PP 
2

= 
4 N 2  i  j 1  Pi
 

i j
 Y 2


Pj  1 P i

1  N  Yi 2 (1  Pi ) 2 PP Y j2 (1  Pi ) 2 PP (1  Pi 2 ) PP
i j 

=   
4 N 2  i  j 1  Pi 2
i j

1  Pi

Pj2
i j

1  Pi
 2YY
i j

  Y
2

PPi j 1 P 
i 

1  N  Y 2 (1  P ) 2 Pj Y 2 (1  Pi ) 2 Pi 
  i  j  2YY
i j (1  Pi )   Y .
i 2
=
4N 2  i  j 1  Pi 1  Pi P j 1  Pi 

13
Using the property
N N N 
 a b
i j   ai   b j  bi  , we can write
i  j 1 i 1  j 1 
1  N Yi 2 (1  Pi ) 2  N  N  N Y j2 Yi 2  N N
Var ( z )    Pj  Pi    Pi (1  Pi )     2 Yi (1  Pi )( Y j  Yi )]  Y
2

4N 2  i 1 Pi (1  Pi )  j 1  i 1  j 1 Pj Pi  i 1 j 1

1  N Yi 2 N  N Y j2 Yi 2  N N 
 2  
4 N  i 1 Pi
(1  Pi
2
 2 Pi )   Pi (1  Pi )      2  Yi (1  Pi )(  Y j  Yi )  Y
2

i 1  j 1 Pj Pi  i 1 j 1

1  N Yi 2 N 2 N N N Y2 N N N Y2

2 
   Yi P  2 Yi   Pi    Yi   Pi  j
2 j 2 2

4 N  i 1 Pi i 1 i 1 i 1 j 1 Pj i 1 i 1 j 1 Pj

N N N N N N
  PY
i i  2 Yi  Y j  2 Yi Pi  2 Yi Pi  Y j  2 Yi ]  Y
2 2 2 2

i i 1 j 1 i 1 i 1 j 1 i 1

1  N 2
Yi N Y N 2
 N N
  2  Pi    Yi  2Ytot  2Ytot  Yi Pi   Y
2 2 2 j 2

4N 2  i 1 i i 1
P j 1 P j i 1 i 1 
 1 N  1  N Yi 2 2 2  1 N 2 N

 2 1   Pi 2  2   Ytot  Ytot   2   Yi  2Y 2
tot  2Ytot  Yi Pi  4 N 2Y 2 
 2 i 1  4 N  i 1 Pi  4 N  i 1 i 1 
2
 1 N  1 N
Y  1 N N
  1   Pi 2  2  Pi  i  Ytot   2  i
( Y 2  2Ytot  Yi Pi  2Ytot2  4Ytot2 )
 2 i 1  2 N i 1  Pi  4 N i 1 i 1

 1 N  1
  1   Pi 2  Y2
2 tot
 2 i 1  2 N
2
 1 N  1 N
Y  1 N N
 1   Pi 2  2  Pi  i  Ytot   2
(  Yi
2
 2Ytot  Yi Pi  2Ytot2  2Ytot2   Pi 2Ytot2 )
 2 i 1  2 N i 1  i
P  4 N i 1 i 1 i
2
 1 N  1 Y 
 Y  2YtotYi Pi  Pi 2Ytot2 
N N
1
 1   Pi 2  2  Pi  i  Ytot   2 i
2

 2 i 1  2 N i 1  Pi  4N i 1
2
1  1 N 2  N  Yi  1 N
Y 
 2 
1   Pi   Pi   Ytot   2  Pi  i  Ytot 
2
`
2 N  2 i 1  i 1  Pi  4N i 1  Pi 
2 2 2
1 N  Y  1 N
 Yi N
 1 N 2  Yi 
  Pi  i  Y   2  Pi 
2
 Ytot   2  i 
P  Ytot 
2 i 1  NPi  4N i 1 i 1  Pi  4 N i 1  Pi 
 
variance of WR reduction of variance
case for n  2 in WR with varying
probability

14
Estimation of Var ( z )
Var ( z )  E ( z 2 )  ( E ( z )) 2
 E(z 2 )  Y 2
Since
E ( z1 z2 )  E  z1 E ( z2 | u1 )
 E  z1Y 
 YE ( z1 )
 Y 2.
Consider
E  z 2  z1 z2   E ( z 2 )  E ( z1 z2 )
 E(z 2 )  Y 2
 Var ( z )
 ( z )  z 2  z z is an unbiased estimator of Var ( z )
 Var 1 2

Alternative form
(z )  z 2  z z
Var 1 2

z z 
2

  1 2   z1 z2
 2 
( z1  z2 ) 2

4
2
1  y1 y y 1  p1 
   1 2 
4  Np1 N N p2 
2
1  y1 y2 (1  p1 ) 
  (1  p1 )  
4N 2  p1 p2 
2
(1  p1 ) 2  y1 y2 
    .
4 N 2  p1 p2 

Case 2: General Case

Let (U i (1) , U i (2) ,..., U i ( r ) ,..., U i ( n ) ) be the units selected in the order in which they are drawn in n draws

where U i ( r ) denotes that the ith unit is drawn at the rth draw. Let ( y1 , y2 ,.., yr ,..., yn ) and

( p1 , p2 ,..., pr ,..., pn ) be the values of study variable and corresponding initial probabilities of selection,

respectively. Further, let Pi (1) , Pi (2) ,..., Pi ( r ) ,..., Pi ( n ) be the initial probabilities of

U i (1) , U i (2) ,..., U i ( r ) ,..., U i ( n ) , respectively.

15
Further, let
y1
z1 
Np1

1  yr 
zr   y1  y2  ...  yr 1  (1  p1  ...  pr 1 )  for r  2,3,..., n.
N  pr 

1 n
Consider z   zr as an estimator of population mean Y .
n r 1

We already have shown in case 1 that E ( z1 )  Y .

Now we consider E ( zr ), r  2,3,..., n. We can write

1
E ( zr )  E1 E2  zr U i (1) , U i (2) ,..., U i ( r 1) 
N
where E2 is the conditional expectation after fixing the units U i (1) ,U i (2) ,..., U i ( r 1) drawn in the first (r -

1) draws.
Consider
y  y 
E  r (1  p  ...  p )   E E  r (1  p  ...  p ) U ,U ,...,U 
p 1 r 1  1 2p 1 r  1 i (1) i(2) i (r  1) 
 r   r 
 y 
 E (1  P  P ...  P ) E  r U ,U ,...,U  .
1 i(1) i (2) i(r  1) 2  p i(1) i(2) i (r  1)  
  r 

y Y
Since conditionally r can take any one of the ( N - r -1) values j , j  1, 2,..., N with probabilities
p P
r j
P
j
, so
1  P  P ...  P
i(1) i (2) i(r  1)

y   N Yj P 
j
E  r (1  p  ...  p )   E (1  P  P ...  P )  * . 
p 1 r 1  1 i(1) i(2) i(r  1) P (1  P  P ...  P )
 r   j  1 j i(1) i(2) i(r  1) 
 N 
 E   *Y 
1 j
 j 1 
N
where  * denotes that the summation is taken over all the values of y except the y values selected in the first (r -1) draws
j 1
N
like as  , i.e., except the values y , y ,..., y
1 2 r 1
which are selected in the first (r -1) draws.
j  1( i(1), i(2),..., i(r  1))

16
Thus now we can express

1  y 
E ( zr )  E1E2  y1  y2  ...  yr 1  r (1  p1  ...  pr 1 ) 
N  pr 
1  N 
 E1 Yi (1)  Yi (2)  ...  Yi ( r 1)   *Y j 
N  j 1 
1  N 
 E1 Yi (1)  Yi (2)  ...  Yi ( r 1) 
N   Yj 
j 1( i (1),i (2),...,i ( r 1)) 


1 
 
E1 Yi (1)  Yi (2)  ...  Yi ( r 1)  Ytot  Yi (1)  Yi (2)  ...  Yi ( r 1) 
N  

1
 E Y 
N 1  tot 
Y
 tot
N
 Y for all r  1, 2,..., n.

Then
1 n
Ez    E  zr 
n r 1
1 n
 Y
n r 1
Y.
Thus z is an unbiased estimator of population mean Y .
The expression for variance of z in general case is complex but its estimate is simple.
Estimate of variance:
Var ( z )  E ( z 2 )  Y 2 .
Consider for r  s,

E ( zr zs )  E  zr E ( zs | U1 , U 2 ,..., U s 1 )

 E  zrY 

 YE ( zr )
Y2

because for r  s, zr will not contribute

and similarly for s  r , zs will not contribute in the expectation.

17
Further, for s  r ,

E ( zr zs )  E  zs E ( zr | U1 , U 2 ,..., U r 1 )

 E  zsY 

 YE ( zs )
 Y 2.
Consider
 1 n n  1 n n
E   r s  n(n  1) r (
z z   E ( zr z s )
 n(n  1) r (  s ) 1 s 1   s ) 1 s 1

1
 n(n  1)Y 2
n(n  1)
 Y 2.
Substituting Y 2 in Var ( z ), we get

Var ( z ) = E ( z 2 )  Y 2
 1 n n 
= E(z 2 )  E    E ( zr zs ) 
 n(n  1) r (  s ) 1 s 1 
n n
(z )  z 2  1
 Var   zr z s .
n(n  1) r (  s ) 1 s 1
2
 n  n n n
Using   zr    zr2    zr zs
 r 1  r 1 r (  s ) 1 s 1
n n n
  z z
r (  s ) 1 s 1
r s  n 2 z 2   zr2 ,
r 1

 ( z ) can be further simplified as

The expression of Var

1  2 2 n 2
(z )  z 2 
Var n z   zr 
n(n  1)  r 1 
1  n 2 
  
n(n  1)  r 1
zr  nz 2 

n
1
  ( zr  z ) 2 .
n(n  1) r 1

18
Unordered estimator:
In ordered estimator, the order in which the units are drawn is considered. Corresponding to any ordered
estimator, there exist an unordered estimator which does not depend on the order in which the units are
drawn and has smaller variance than the ordered estimator.
N
In case of sampling WOR from a population of size N , there are   unordered sample(s) of size n .
n
Corresponding to any unordered sample(s) of size n units, there are n ! ordered samples.
For example, for n  2 if the units are u1 and u2 , then

- there are 2! ordered samples - (u1 , u2 ) and (u2 , u1 )

- there is one unordered sample (u1 , u2 ) .

Moreover,
 Probability of unordered   Probability of ordered   Probability of ordered 
    
 sample (u1 , u2 )   sample (u1 , u 2 )   sample (u2 , u 1 ) 
For n  3, there are three units u1 , u2 , u3 and

-there are following 3! = 6 ordered samples:

(u1 , u2 , u3 ), (u1 , u3 , u2 ), (u2 , u1 , u3 ), (u2 , u3 , u1 ), (u3 , u1 , u2 ), (u3 , u2 , u1 )

- there is one unordered sample (u1 , u2 , u3 ).

Moreover,
Probability of unordered sample
= Sum of probability of ordered sample, i.e.
P(u1 , u2 , u3 )  P (u1 , u3 , u2 )  P (u2 , u1 , u3 )  P (u2 , u3 , u1 )  P (u3 , u1 , u2 )  P (u3 , u2 , u1 ),

N
Let zsi , s  1, 2,..,   , i  1, 2,..., n !( M ) be an estimator of population parameter  based on ordered
n
sample si . Consider a scheme of selection in which the probability of selecting the ordered sample

( si ) is psi . The probability of getting the unordered sample(s) is the sum of the probabilities, i.e.,
M
ps   psi .
i 1

For a population of size N with units denoted as 1, 2,…, N , the samples of size n are n  tuples. In the
nth draw, the sample space will consist of N ( N  1)...( N  n  1) unordered sample points.

19
 1
psio  P selection of any ordered sample 
N ( N  1)...( N  n  1)
n! selection of any 
psiu  P selection of any unordered sample   n!P  
N ( N  1)...( N  n  1)  ordered sample 
M (  n!)
n !( N  n)! 1
then ps  
i 1
psio 
N!

N
.
 
n

N
Theorem : ˆ0  zsi , s  1, 2,...,   ; i  1, 2,..., M ( n !)
n
M
and ˆu   zsi psi'
i 1

are the ordered and unordered estimators of  , then

(i) E (ˆu )  E (ˆ0 )

(ii) Var (ˆu )  Var (ˆ0 )

where zsi is a function of si th ordered sample (hence a random variable) and psi is the probability of

psi
selection of si th ordered sample and ps' i  .
ps

N
Proof: Total number of ordered sample = n ! 
n
N
 
n M
(i ) E (ˆ0 )   zsi psi
s 1 i 1
N
 
n
M 
E (ˆu )     zsi psi'  ps
s 1  i 1 
 p 
    zsi si  ps
s  i ps 
  zsi psi
s i

 E (ˆ0 )

N
(ii) Since ˆ0  zsi , so ˆ02  zsi2 with probability psi , i  1, 2,..., M , s  1, 2,...,   .
n
2
M
M 
Similarly, ˆu   zsi psi' , so ˆu2    zsi psi'  with probability ps
i 1  i 1 

20
Consider
2
Var (ˆ0 )  E (ˆ02 )   E (ˆ0 ) 

  zsi2 psi   E (ˆ0 ) 

2

s i
2
Var (ˆu )  E (ˆ )   E (ˆu ) 
2
u

2
 
    zsi psi'  ps   E (ˆ0 ) 
2

s  i 
2
 
Var (ˆ0 )  Var (ˆu )   zsi2 psi     zsi psi'  ps
s i s  i 
2
 
  zsi2 psi     zsi psi  ps
s i s  i 
  
 2   zsi psi'   zsi psi  ps
s  i  i 
 
2
      
    zsi2 psi    zsi psi'    psi   2   zsi psi'   zsi psi  ps 
s   i  i   i   i  i  
  
2

'   
     zsi psi    zsi psi  psi  2   zsi psi'  zsi psi 
2

s  i   i   i  
 
 
  ( zsi   zsi psi' ) 2 psi   0
s i  i 
 Var (ˆ0 )  Var (ˆu )  0
or Var (ˆ )  Var (ˆ )
u 0

Estimate of Var (ˆu )

Since
 
Var (ˆ0 )  Var (ˆu )   ( zsi   zsi psi' ) 2 psi 
s i  i 
 
 (ˆ )  Var
Var u
 (ˆ ) 
0 
s i 
( zsi   zsi psi ) psi 
i
' 2


 (ˆ )  p ' ( 
 
p ' Var
i
si 0 
z  z p ' )2 .
i
si si  i
si si

Based on this result, now we use the ordered estimators to construct an unordered estimator. It follows
from this theorem that the unordered estimator will be more efficient than the corresponding ordered
estimators.

21
Murthy’s unordered estimator corresponding to Des Raj’s ordered estimator for the
sample size 2
Suppose yi and y j are the values of units U i and U j selected in the first and second draws respectively

with varying probability and WOR in a sample of size 2 and let pi and p j be the corresponding initial

probabilities of selection. So now we have two ordered estimates corresponding to the ordered samples
s1* and s2* as follows

s1*  ( yi , y j ) with (U i , U j )
s2*  ( y j , yi ) with (U j , U i )

which are given as

1  yi yj 
z ( s1* )   (1  pi )  (1  pi ) 
2N  pi p j 

where the corresponding Des Raj estimator is given by

1  yi y j (1  pi ) 
 yi   
2N  pi pj 
and

1  yj yi 
z ( s2* )  (1  p j )  (1  p j ) 
2N  pj pi 

where the corresponding Des Raj estimator is given by

1  y j yi (1  p j ) 
yj   .
2N  pj pi 

The probabilities corresponding to z ( s1* ) and z ( s2* ) are

pi p j
p ( s1* ) 
1  pi
p j pi
p ( s2* ) 
1 p j

p ( s )  p ( s1* )  p ( s2* )
pi p j (2  pi  p j )

(1  pi )(1  p j )

22
 1 p j
p '( s1* ) 
2  pi  p j
1  pi
p '( s2* )  .
2  pi  p j

Murthy’s unordered estimate z (u ) corresponding to the Des Raj’s ordered estimate is given as

z (u )  z ( s1* ) p '( s1 )  z ( s2* ) p '( s2 )

z ( s1* ) p ( s1* )  z ( s2* ) p ( s2* )

p ( s1* )  p ( s2* )
 1  yi y j   pi p j    1  yj yi  p j pi 
 (1  pi )  (1  pi )      (1  p j )  (1  p j )    
 2N  pi p j   1  pi    2 N  pj pi  1  p j  

pi p j p p
 j i
1  pi 1  p j

1   y y j   yj y  
 (1  pi ) i  (1  pi )  (1  p j )  (1  p j )  (1  p j ) i  (1  pi ) 

2N   pi p j   pi pi  
(1  p j )  (1  pi )

1  
(1  p j ) (1  pi )  (1  pi )  (1  pi ) (1  p j )  (1  p j 
yi yj
2 N  pi pj 

2  pi  p j

yi y
(1  p j )  (1  pi ) j
pi pj
 .
N (2  pi  p j )

Unbiasedness:
Note that yi and pi can take any one of the values out of Y1 , Y2 ,..., YN and P1 , P2 ,..., PN ,

respectively. Then y j and p j can take any one of the remaining values out of Y1 , Y2 ,..., YN and

P1 , P2 ,..., PN , respectively, i.e., all the values except the values taken at the first draw. Now

23
   Y Y j   PP PP 
 (1  Pj ) i  (1  Pi )    i j 
i j

  Pi Pj  1  Pi 1  Pj 
E  z (u )    
1
N i j 2  Pi  Pj

  Y Y j   PP P P 
 (1  Pj ) i  (1  Pi )    j i 
i j

  Pi Pj  1  Pi 1  Pj 
2 
1

2 N i j 2  Pi  Pj

  Y Y j   PP P P 
 (1  Pj ) i  (1  Pi )    j i 
i j

1   Pi Pj  1  Pi 1  Pj 
 
2 N i j 2  Pi  Pj

1   Yi Y j   PP 
   (1  P ) P  (1  P ) P   (1  P )(1  P ) 
j i
i j

2N i j   i j   i j 

1  Yi Pj Y j Pi 
  1  P  1  P 
2N i j  i j 
N N  N 
Using result  a b   a  b
i j i j  bi , we have
i  j 1 i 1  j 1 

1  N Y N   N Y j N 
E  z (u )     i ( Pj  Pi )    ( Pi Pj ) 
2N   i 1 1  Pi j 1   j 1 1  Pj i 1 

1   N Yi  N Y 
   (1  Pi )    j (1  Pj ) 
2N   i 1 1  Pi  j 1 1  Pj 

1 N N 
  Yi   Y j 
2 N  i 1 j 1 

Y Y

2
Y.

24
Variance: The variance of z (u ) can be found as
2
1 N (1  Pi  Pj )(1  Pi )(1  Pj )  Yi Y j  PP
i j (2  Pi  Pj )
Var  z (u )      
2 i  j 1 N 2 (2  Pi  Pj )  P P  (1  P )(1  P )
 i j  i j
2
1 N PP (1  Pi  Pj )  Yi Y j 
  i 2j   
2 i  j 1 N (2  Pi  Pj )  Pi Pj 

Using the theorem that Var (ˆu )  Var (ˆ0 ) we get

Var  z (u )   Var  z ( s1* ) 

and Var  z (u )   Var  z ( s2* ) 

Unbiased estimator of V  z (u )

An unbiased estimator of Var  z | u  is

2

  z (u )   (1  pi  p j )(1  pi )(1  p j )  yi  y j  .
Var
N 2 (2  pi  p j ) 2 p p 
 i j 

Horvitz Thompson (HT) estimate

The unordered estimates have limited applicability as they lack simplicity and the expressions for the
estimators and their variance becomes unmanageable when sample size is even moderately large. The
HT estimate is simpler than other estimators. Let N be the population size and yi , (i  1, 2,..., N ) be the

value of characteristic under study and a sample of size n is drawn by WOR using arbitrary probability
of selection at each draw.

Thus prior to each succeeding draw, there is defined a new probability distribution for the units available
at that draw. The probability distribution at each draw may or may not depend upon the initial
probability at the first draw.

Define a random variable  i (i  1, 2,.., N ) as

1 if Yi is included in a sample ' s ' of size n

i  
0 otherwise.

25
 nyi
Let zi  , i  1...N assuming E ( i )  0 for all i
NE ( i )
where
E ( i )  1.P (Yi  s )  0.P(Yi  s )
 i

is the probability of including the unit i in the sample and is called as inclusion probability.

The HT estimator of Y based on y1 , y2 ,..., yn is

1 n
zn  YˆHT   zi
n i 1
1 N
   i zi .
n i 1

Unbiasedness
1 N
E (YˆHT )   E ( zi i )
n i 1
1 N
  zi E ( i )
n i 1
1 N nyi
  E ( i )
n i 1 NE ( i )
1 N nyi
  Y
n i 1 N
which shows that HT estimator is an unbiased estimator of population mean.

Variance
V (YˆHT )  V ( zn )
 E ( zn2)   E ( zn ) 
2

 E ( zn2)  Y 2 .
Consider
2
1 N 
E ( z )  2 E    i zi 
n
2

n  i 1 
1  N N N 
 2 E    i2 zi2     i j zi z j 
n  i 1 i (  j ) 1 j 1 
1  N N N 
 2   zi2 E ( i2 )    zi z j E ( i j )  .
n  i 1 i (  j ) 1 j 1 

26
If S  s is the set of all possible samples and  i is probability of selection of ith unit in the sample s

then
E ( i )  1 P( yi  s )  0.P ( yi  s )
 1. i  0.(1   i )   i
E ( )  12. P( yi  s )  02.P( yi  s )
i
2

 i.
So
E ( i )  E ( i2 )
 
1 N 2 N N
E ( z )  2  zi  i     ij zi z j 
2

n  i 1 
n
i (# j ) i 1
 

where  ij is the probability of inclusion of ith and jth unit in the sample. This is called as second order

inclusion probability.
Now

Y 2   E ( zn ) 
2

2
1   N 
 2  E    i zi  
n   i 1 

1 N 2 2
N N
 i z  E ( )     zi z j E ( i ) E ( j )

n 2  i 1
i
 i (  j ) 1 j 1

1 N 2 2 N N 
2  i i  
 z    i j zi z j  .
n  i 1 i (  j ) 1 j 1 

Thus

1 N N N 
Var (YˆHT )  2    i zi2     ij zi z j 
n  i 1 i (  j ) 1 j 1 
1  N N N 
 2    i2 zi2     i j zi z j 
n  i 1 i (  j ) 1 j 1 
1 N N N 
 2    i (1   i ) zi2    ( ij   i i ) zi z j 
n  i 1 i (  j ) 1 j 1 
1 N n 2 yi2 N N n 2 yi y j 
  i  (1   )    ij i i N 2  
(    )
N 2 i2 i (  j ) 1 j 1
i
n 2  i 1 i j 
1  N  1   2 N N      
        yi y j 
ij i i
 i
i
y
N2 
 i 1  i  i (  j ) 1 j 1   
i j  
27
Estimate of variance
1  n yi2 (1   i ) n n      yi y j 
V1  Var (YHT )  2  
ˆ  ˆ     ij i j
 .
i  i j   i j 
2
N  i 1 i (  j ) 1 j 1 

This is an unbiased estimator of variance .

yi
Drawback: It does not reduces to zero when all are same, i.e., when yi   i .
i
Consequently, this may assume negative values for some samples.
A more elegant expression for the variance of yˆ HT has been obtained by Yates and Grundy.

Yates and Grundy form of variance

Since there are exactly n values of  i which are 1 and ( N  n) values which are zero, so
N


i 1
i  n.

Taking expectation on both sides

N

 E ( )  n.
i 1
i

Also
2
 N  N N N
E    i    E ( i2 )    E ( i j )
 i 1  i 1 i (  j ) 1 j 1
N N N
E  n    E ( i )    E (  ) (using E ( i )  E ( i2 ))
2
i J
i 1 i (  j ) 1 j 1
N N
n2  n    E ( 
i (  j ) 1 j 1
i J )

N N

  E ( 
i (  j ) 1 j 1
i J )  n(n  1)

Thus E ( i j )  P( i  1,  j  1)
 P ( i  1) P ( j  1  i  1)
 E ( i ) E ( j  i  1)

28
Therefore
N


j (  i ) 1
 E ( i  j )  E ( i ) E ( j ) 

N
 
j (  i ) 1
 E ( i ) E ( j |  i  1)  E ( i ) E ( j ) 

N
 E ( i ) 
j (  i ) 1
 E ( j |  i  1)  E ( j ) 

 E ( i )  (n  1)  (n  E ( i )
  E ( i ) 1  E ( i ) 
  i (1   i ) (1)

Similarly
N


i (  j ) 1
 E ( i  j )  E ( i ) E ( j )    j (1   j ). (2)

We had earlier derived the variance of HT estimator as

1 N N N 
Var (YˆHT )  2    i (1   i ) zi2    ( ij   i j ) zi z j 
n  i 1 i (  j ) 1 j 1 
Using (1) and (2) in this expression, we get
1 N N N N 
Var (YˆHT )  2    i (1   i ) zi2    j (1   j ) z 2j  2   ( i j   ij ) z i z j 
2n  i 1 j 1 i  j 1 j 1 
1  N  N 
 2     E ( i j )  E ( i ) E ( j )  zi2
2n  i 1  j ( i ) 1 
N  N  
    E ( i j )  E ( i ) E ( j )  z 2j  2    E ( i ) E ( j )  E ( i j ) zi z j 
N n

j 1 i (  j ) 1  i (  j ) 1 j 1 

 N N
1 N N N N 
     ( ij   i i ) zi    ( ij   i i ) z j  2   ( ij   i i ) zi z j 
2 2

2n 2
  i (  j ) 1 j 1 i (  j ) 1 j 1 i (  j ) 1 j 1 
1  N N 
 2    ( i j   ij )( zi2  z 2j  2 zi z j )  .
2n  i (  j ) 1 j 1 

The expression for  i and  ij can be written for any given sample size.

For example, for n  2 , assume that at the second draw, the probability of selecting a unit from the units
available is proportional to the probability of selecting it at the first draw. Since

29
E ( i )  Probability of selecting Yi in a sample of two

 Pi1  Pi 2

where Pir is the probability of selecting Yi at r th draw (r  1, 2). If Pi is the probability of selecting the

ith unit at first draw (i  1, 2,..., N ) then we had earlier derived that
Pi1  Pi
 yi is not selected   yi is selected at 2nd draw| 
Pi 2  P  st P 
  yi is not selected at 1 draw 
st
at 1 draw
N PP
  j i
j (  i ) 1 1  Pj

N P P 
   j  i  Pi .
 j 1 1  Pj 1  Pi 
So
N P P 
E ( i )  Pi   j  i 
 j 1 1  Pj 1  Pi 
Again
E ( i j )  Probability of including both yi and y j in a sample of size two
 Pi1 Pj 2|i  Pj1 Pi 2| j
Pj Pi
 Pi  Pj
1  Pi 1  Pj
 1 1 
i j 
=PP  .
1  Pi 1  Pj 

Estimate of Variance
The estimate of variance is given by

 (Yˆ )  1
n n  i j   ij
Var HT
2n 2

i( j ) j 1  ij
( zi z j ) 2 .

30
Midzuno system of sampling:
Under this system of selection of probabilities, the unit in the first draw is selected with unequal
probabilities of selection (i.e., pps) and remaining all the units are selected with SRSWOR at all
subsequent draws.

Under this system

E ( i )   i  P (unit i (U i ) is included in the sample)

 P (U i is included in 1st draw) + P (U i is included in any other draw )

 Probability that U i is not selected at the first draw and 

 Pi   
 is selected at any of subsequent (n -1) draws 
 
n 1
 Pi  (1  Pi )
N 1
N n n 1
 Pi  .
N 1 N 1
Similarly,
E ( i j )  Probability that both the units U i and U j are in the sample

 Probability that U i is selected at the first draw and 

  
 U is selected at any of the subsequent draws (n  1) draws 
 j 

 Probability that U j is selected at the first draw and

  
 U is selected at any of the subsequent ( n  1) draws 
 i 
 Probability that neither U i nor U j is selected at the first draw but 
  
 both of them are selected during the subsequent (n  1) draws 
 
n 1 n 1 (n  1)(n  2)
 Pi  Pj  (1  Pi  Pj )
N 1 N 1 ( N  1)( N  2)

(n  1)  N  n n2 
  ( Pi  Pj ) 
( N  1)  N  2 N  2 

n 1  N  n n2 
 ij   ( Pi  Pj )  .
N 1  N  2 N  2 
Similarly,
E ( i j k )   ijk  Probability of including U i , U j and U k in the sample
(n  1)(n  2)  N  n n3 
  ( Pi  Pj  Pk )  .
( N  1)( N  2)  N  3 N  3 

31
By an extension of this argument, if U i , U j ,..., U r are the r units in the sample of size n(r  n), the

probability of including these r units in the sample is

(n  1)(n  2)...(n  r  1)  N  n nr 
E ( i j ... r )   ij ...r   ( Pi  Pj  ...  Pr ) 
( N  1)( N  2)...( N  r  1)  N  r N  r 
Similarly, if U1 , U 2 ,..., U q be the n units, the probability of including these units in the sample is

(n  1)(n  2)...1
E ( i j ... q )   ij ...q  ( Pi  Pj  ...  Pq )
( N  1)( N  2)...( N  n  1)
1
 ( Pi  Pj  ...  Pq )
 N  1
 
 n 1 
which is obtained by substituting r  n .

Thus if Pi ' s are proportional to some measure of size of units in the population then the probability of

selecting a specified sample is proportional to the total measure of the size of units included in the
sample.
Substituting these  i ,  ij ,  ijk etc. in the HT estimator, we can obtain the estimator of population’s mean

and variance. In particular, an unbiased estimate of variance of HT estimator given by

 (Yˆ )  1
n n  i j   ij
Var HT
2n 2

i  j 1 j 1 
( zi  z j ) 2
ij

where
N n  n 1 
 i j   ij  ( N  n) PP
i j  (1  Pi  Pj )  .
( N  1) 2 N 2 

The main advantage of this method of sampling is that it is possible to compute a set of revised
probabilities of selection such that the inclusion probabilities resulting from the revised probabilities are
proportional to the initial probabilities of selection. It is desirable to do so since the initial probabilities
can be chosen proportional to some measure of size.

32