SAMPLING DISTRIBUTION

Introduction
• In real life calculating parameters of populations is
usually impossible because populations are very
large.
• Rather than investigating the whole population,
we take a sample, calculate a statistic related to
the parameter of interest, and make an inference.
• The sampling distribution of the statistic is the
tool that tells us how close is the statistic to the
parameter.

2
 Sampling Distribution of the Mean
• An example
– A die is thrown infinitely many times. Let X
represent the number of spots showing on any
throw.
– The probability distribution of X is

E(X) = 1(1/6) +
x 1 2 3 4 5 6 2(1/6) + 3(1/6)+
………………….= 3.5
p(x) 1/6 1/6 1/6 1/6 1/6 1/6
V(X) = (1-3.5)2(1/6) +
(2-3.5)2(1/6) +
3
 Throwing a die twice – sample mean

• Suppose we want to estimate  from

the mean X of a sample of size n = 2.
• What is the distribution of X?

4
 Throwing a die twice – sample mean

Sample Mean Sample Mean Sample Mean

1 1,1 1 13 3,1 2 25 5,1 3
2 1,2 1.5 14 3,2 2.5 26 5,2 3.5
3 1,3 2 15 3,3 3 27 5,3 4
4 1,4 2.5 16 3,4 3.5 28 5,4 4.5
5 1,5 3 17 3,5 4 29 5,5 5
6 1,6 3.5 18 3,6 4.5 30 5,6 5.5
7 2,1 1.5 19 4,1 2.5 31 6,1 3.5
8 2,2 2 20 4,2 3 32 6,2 4
9 2,3 2.5 21 4,3 3.5 33 6,3 4.5
10 2,4 3 22 4,4 4 34 6,4 5
11 2,5 3.5 23 4,5 4.5 35 6,5 5.5
12 2,6 4 24 4,6 5 36 6,6 6

5
 Sample Mean Sample Mean Sample Mean

The distribution
1
1,1
2
1,2 of 1
1.5
13
14
3,1
3,2 X
2
2.5 when n =5,1
25
262
5,2
3
3.5
3 1,3 2 15 3,3 3 27 5,3 4
4 1,4 2.5 16 3,4 3.5 28 5,4 4.5
5 1,5 3 17 3,5 4 29 2
5,5 5
6 1,6 3.5 18 3,6 4.5 30 2 x
5,6 5.5
7 Note :   
2,1 1.5
x
19 4,1
x
2.5 and   31
32x
6,1 3.5
8
9
2,2
2,3
2
2.5
20
21
4,2
4,3
3
3.5 33
6,2
6,3 2 4
4.5
10 2,4 3 22 4,4 4 34 6,4 5
11 2,5 3.5 23 4,5 4.5 35 6,5 5.5
12 2,6 4 24 4,6 5 36 6,6 6

E( x) =1.0(1/36)+
6/36 1.5(2/36)+….=3.5
5/36

4/36 V(X) = (1.0-3.5)2(1/36)+

(1.5-3.5)2(2/36)... = 1.46
3/36
2/36
1/36
1 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
x 6
Sampling Distribution of the Mean
n5 n  10 n  25
 x  3.5  x  3.5  x  3.5
2  2x  2x  2x
  .5833 (  )
x
2
  .2917 (  ) 2
  .1167 (  )
5 x
10
x
25
6

7
Sampling Distribution of the Mean

n5 n  10 n  25
 x  3.5  x  3.5  x  3.5
 2x 2  2x 2  2x
2
 x  .5833 (  )   .2917 (  )
x   .1167 (  )
x
5 10 25

Notice
Noticethat that x2 is
is smaller than .x.2
smaller than
Thelarger
The largerthethesample
samplesize sizethe
the
Therefore,x xtends
2
smaller 
smaller x .. Therefore, tends
totofall
fallcloser
closertoto,,asasthe
thesample
sample
sizeincreases.
size increases. 8
 SAMPLING DISTRIBUTION
• Let X1, X2,…,Xn be a r.s. of size n from a
population and let T(x1,x2,…,xn) be a real (or
vector-valued) function whose domain includes
the sample space of (X1, X2,…,Xn). Then, the
r.v. or a random vector Y=T(X1, X2,…,Xn) is
called a statistic. The probability distribution of
a statistic Y is called the sampling distribution
of Y.

9
 SAMPLING DISTRIBUTION
• The sample mean is the arithmetic average of
the values in a r.s.
X1  X 2    X n 1 n
X   Xi
n n i 1
• The sample variance is the statistic defined by
1 n
 Xi  X 
2
S 
2

n  1 i1
• The sample standard deviation is the statistic
defined by S.
10
 SAMPLING FROM THE NORMAL
DISTRIBUTION
Properties of the Sample Mean and Sample
Variance
• Let X1, X2,…,Xn be a r.s. of size n from a
N(,2) distribution. Then,
a ) X and S are independent rvs.
2

b) X ~ N   ,  / n 
2

c)
 n  1 S
2

~  n1
2

 2

11
 SAMPLING FROM THE NORMAL
DISTRIBUTION
• Let X1, X2,…,Xn be a r.s. of size n from a
N(,2) distribution. Then,
X 
~ N  0,1
/ n
•Most of the time  is unknown, so we use:

X 
.
S/ n
12
 SAMPLING FROM THE NORMAL
DISTRIBUTION

In statistical inference, Student’s t

distribution is very important.

13
 SAMPLING FROM THE NORMAL
DISTRIBUTION
• Let X1, X2,…,Xn be a r.s. of size n from a
N(X,X2) distribution and let Y1,Y2,…,Ym
be a r.s. of size m from an independent
N(Y,Y2).
• If we are interested in comparing the
variability of the populations, one
quantity of2 interest would be the ratio
 X /  Y  S X / SY
2 2 2

14
 SAMPLING FROM THE NORMAL
DISTRIBUTION
• The F distribution allows us to compare these
quantities by giving the distribution of

S X2 / SY2 S X2 /  X2
 2 2 ~ Fn1,m1
 X /  Y SY /  Y
2 2

• If X~Fp,q, then 1/X~Fq,p.

• If X~tq, then X2~F1,q.

15
 CENTRAL LIMIT THEOREM
If a random sample is drawn from any population, the sampling
distribution of the sample mean is approximately normal for
a sufficiently large sample size. The larger the sample size,
the more closely the sampling distribution of X will resemble
a normal distribution.

Random Sample
(X1, X2, X3, …,Xn)

X X
as n  
Random Variable Sample Mean
(Population) Distribution Distribution
16
Sampling Distribution of the Sample
Mean

X  
 2

X 
2
or  X 
n n
If X is normal, X is normal. If X is non-normal,
X is approximately normally distributed for
sample size greater than or equal to 30.
X 
X ~ N(  , / n )  Z 
2
~ N( 0,1 )
/ n 17
 EXAMPLE 1
• The amount of soda pop in each bottle is normally
distributed with a mean of 32.2 ounces and a
standard deviation of 0.3 ounces.
– Find the probability that a bottle bought by a customer
will contain more than 32 ounces.
– Solution
• The random variable X is the
amount of soda in a bottle. 0.7486

x   32  32.2
P( x  32)  P(  )
x .3 x = 32  = 32.2
 P( z  .67 )  0.7486 18
 EXAMPLE 1 (contd.)

• Find the probability that a carton of four bottles will have

a mean of more than 32 ounces of soda per bottle.
• Solution
– Define the random variable as the mean amount of soda per
bottle.

x   32  32.2
P( x  32)  P(  ) 0.9082
x .3 4
 P( z  1.33 )  0.9082
0.7486
x = 32
x  32  = 32.2
 x  32.2 19
 Sampling Distribution of
a Proportion

• The parameter of interest for nominal data is

the proportion of times a particular outcome
(success) occurs.
• To estimate the population proportion ‘p’ we
use the sample proportion. The number
of successes

^p = X
The estimate of p = n

20
 Sampling Distribution of
a Proportion

• Since X is binomial, probabilities about ^p can

be calculated from the binomial distribution.
^
• Yet, for inference aboutp we prefer to use
normal approximation to the binomial
whenever it approximation is appropriate.

21
 Approximate Sampling Distribution of a
Sample Proportion

• From the laws of expected value and variance, it can be

shown that E( p̂) = p and V( p̂)=p(1-p)/n
• If both np ≥ 5 and n(1-p) ≥ 5, then

pˆ p
z
p (1  p )
n

• Z is approximately standard normally distributed.

22
 EXAMPLE
– A state representative received 52% of the
votes in the last election.
– One year later the representative wanted to
study his popularity.
– If his popularity has not changed, what is the
probability that more than half of a sample of
300 voters would vote for him?

23
 EXAMPLE (contd.)
Solution
• The number of respondents who prefer the representative is
binomial with n = 300 and p = .52. Thus, np = 300(.52) = 156 and
n(1-p) = 300(1-.52) = 144 (both greater than 5)

 pˆ  p .50  .52 
P ( pˆ  .50)  P    .7549
 p(1  p) n (.52)(1  .52) 300 


24