Lecture #4 Sampling Distribution

Objectives of this lecture:

After reading this unit, you should be able to
• understand about the methods of finding sampling distribution

• understand about distribution function methods

Methods of finding the distribution of function of random

variables
i. Transformation of variable
ii. Moment generating function
iii. Distribution function
iv. Geometric method
v. Inverse technique

How you can drive the distribution function method of random

variables?
Solution: Let X, Y, Z, . . . be random variables with joint density function f(X,Y,Z, . . .)
Let, W = h(X,Y, Z, . . . )
1st step:
Find the distribution function of W
G(W )=P [W ≤w]= P [ X , Y , Z , . . .≤w]
2nd Step:
Find the density function of W
g (W )=Ǵ (W )

Example: Let X have a normal distribution with mean 0 and variance 1, the pdf of standard normal
distribution is
2
x
1 −2
f (x )= e ; −∞ < x< ∞
√2 π
Let W = X 2 . Find the distribution of W.

Sampling Distribution | Basic| Lecture #4

 Lectured by
Subject: STT251: Sampling Technique
Md. Kaderi Kibria

Solution: Given the pdf of standard normal distribution of random variable x

2
x
1 −2
f (x )= e ; −∞ < x< ∞
√2 π
1st step:
Find the distribution function of W
G(W )=P [W ≤w ]
2
= P [ X ≤w]
= P [− √ w≤ X ≤ √ w]
2
√w x
1 −2
=∫ e dx
−√ w √ 2 π
= F ( √ W )−F (−√ W )
G (W )=F ( √W )−F (−√ W )
Since, F (x)=∫ f ( x )dx and f (x )= F́ (x )
x

2nd step:
To find the density function of W
g (W )=Ǵ (W )
d
= [F ( √ W )−F (−√ W )]
dw
d√w d √w
= F́ ( √ w) + F́ (− √ w)
dw dw
1 1
1 −2 1 −2
= F́ ( √ w) W + F́ (−√ w) W
2 2
1 1
1 − 1 −
= f ( √ w) W 2 + f (−√ w) W 2
2 2
w 1 w 1
1 −
2 1
−
2 1 −
2 1
−
= e . W + e . W 2
√2 π 2 √2 π 2
w 1
1 − 2 1 −2
= 2. e . W
√2π 2
w 1
1 − 2 −2
= e W
√2 π
w 1
1 − 2 −2
g (W )= e W ; W ≥0
√2 π
Thus, X has a standard normal distribution, then W = X 2 has the density function
w 1
1 − 2 −2
g (W )= e W ; W ≥0
√2π
1
1 2
( ) −w −1
2
= e 2W 2
1
Γ( )
2
This distribution is the gamma distribution.

Sampling Distribution | Basic| Lecture #4

 Lectured by
Subject: STT251: Sampling Technique
Md. Kaderi Kibria

x−μ
Example: If the random variable X ∼N (μ , σ2 ) then the random variable Z = σ ∼N (0,1).
Solution: We have
X ∼N ( μ , σ2 )
Now, the distribution function of Z is
G (Z )=Pr [ Z ≤z ]
x−μ
= Pr [ σ ≤z ]
= Pr [ x−μ ≤z σ ]
= Pr [ x≤μ + z σ ]
2
μ+ z σ μ+ z σ 1 x−μ
1 − ( σ )
= ∫ f (x )dx= ∫ e 2 dx
−∞ −∞ σ √2π
z 1
1
2
− z
=∫ e 2 σ dz
−∞ σ √ 2 π
z 1
1 −2 z
2

=∫ e dz
−∞ √ 2 π
z 1
1 −2 z
2

G (Z )= ∫ e dz
−∞ √ 2 π

Now, the density function of Z is

d
g ( Z )= [G(Z )]
dz
z −1 2
=
d
dz [∫ −∞
1
√2 π
e 2
z
dz ]
−1 2 z
=
1
√2π
e 2
z

−1 2
| −∞

1 2
z
= e
√2π
−1
1
2
z
g (Z )= e 2 ;−∞ < z < ∞
√2 π
Which is the pdf of standard normal distribution with mean 0 and variance 1.

Example: Suppose that X and Y are independent random variables each having an exponential
distribution with parameter λ
−λ x
f 1 ( X )= λ e ; for x≥0
−λ y
and f 2 (Y )= λ e ; for y≥0
f ( X ,Y )= f 1 ( X ) f 2 (Y )
Also,
= λ 2 e −λ (x+ y) ; x ≥0 , y ≥0
Let W=X+Y. Find the distribution of W.

Sampling Distribution | Basic| Lecture #4

 Lectured by
Subject: STT251: Sampling Technique
Md. Kaderi Kibria

Solution:
Given that
−λ x
f 1 ( X )= λ e ; for x≥0
and f 2 (Y )= λ e−λ y ; for y≥0
f ( X , Y )= f 1 ( X ) f 2 (Y )
Also, 2 − λ (x+ y)
=λ e ; x≥0 , y ≥0
1st Step: Find the distribution function of W
G(W )=P [W ≤w]
= P [ X +Y ≤w]
= P [ X ≤w−Y ]
= P [Y ≤w− X ]
w w−x

=∫ ∫ f 1 ( x) f 2 ( y) dy dx
0 0
w w−x

=∫ ∫ λ2 e−λ (x+ y) dy dx
0 0
w w− x
=λ
2
∫e
0
−λ x
[∫ ]
0
e
−λ y
dy dx
w − λ y w− x
=λ 2

w
0
∫e −λ x
[ | ]
e
−λ 0
dx

−λ x
= λ∫e [−e−λ (w− x) +e 0 ] dx
0
w
−λ x
= λ∫e [ 1−e−λ w+ λ x ] dx
0
w
−λ x −λ x − λw λx
= λ∫[e −e .e .e ] dx
0
w
−λ x − λw
= λ∫[e −e ] dx
0
w
−λ x
=λ
e
[
−λ
−e−λ w x ]| 0
w
−λ x −λ w
=λ [
e +λ x e
−λ 0
]|
=− [ e + λ w e−λ w −e0 ]
−λ w

−λ w −λ w
=− [ e + λ w e −1 ]
−λ w −λ w
= 1−e −λ w e
G(W )=1−e−λ w − λ w e −λ w ; w≥0
Which is the distribution function of W.

2nd Step:
Find the density function of W

Sampling Distribution | Basic| Lecture #4

 Lectured by
Subject: STT251: Sampling Technique
Md. Kaderi Kibria

g (W )=Ǵ(W )
d
= [1−e−λ w − λ w e− λ w ]
dw
−λ w −λ w 2 −λ w
=0−(− λ )e −λ e + λ w e
−λ w −λ w 2 −λ w
= λ e −λ e + λ w e
2 −λ w
=λ w e
2 −λ w
g(W )=λ w e ; w≥0
Hence, the density function of X+Y is
g(W )= λ2 w e−λ w ; w≥0
This distribution function recognized to be the gamma distribution with parameter r=2 and λ
2
g (W )= λ e
−λ w 2−1
w ; w≥0
Γ2

References:
1. Hogg, R. V. and Craig, A. T. (2012), Introduction of Mathematical Statistics, 7 th Edition,
Pearson Education, New Delhi.
2. Hoel, P. G. (1984), Introduction to Mathematical Statistics, 5th Edition, John Wiley and
Sons, New York.
3. Mood, A. M., Graybill, F. A., Boes, D. C. (1974), Introduction to the Theory of Statistics,
3rd Edition, McGraw-Hill, USA.
4. https://people.ohio.edu/ruhil/statsbook/sampling.html#the-sampling-distribution-of-binary-
proportions
5. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://egyankosh.ac.in/bitstream/
123456789/14027/1/Unit-4.pdf

Sampling Distribution | Basic| Lecture #4