MA 4040/ MA 2540: Probability Theory

Dr. Sameen Naqvi

Department of Mathematics, IIT Hyderabad
Email id: [email protected]

1 / 19
 Function of a Discrete Random Variable

I Theorem 2: Let X be a random variable of discrete type with

support SX and p.m.f. fX (·).

Let h : R → R be a Borel function and let Z : Ω → R be

defined by Z (ω) = h (X (ω)) , ω ∈ Ω.

Then Z is a random variable of discrete type with support

SZ = {h (x) : x ∈ SX } and p.m.f.
 P
 fX (x), if z ∈ SZ
fZ (z) = x∈h−1 (z)
0, otherwise


2 / 19
Function of a Discrete Random Variable contd.

I Corollary 1:
Under the notation and assumptions of Theorem 2, suppose
that h : R → R is one-one with inverse function h−1 : D → R,
where D = {h (x) : x ∈ R}.

Then Z is a discrete type random variable with support

Sz = {h (x) : x ∈ SX } and p.m.f.
(
fX (h−1 (z)), if z ∈ SZ
fZ (z) =
0, otherwise

3 / 19
 Example 1

I Let X be a random variable with p.m.f.


1
7,
 if x ∈ {−2, −1, 0, 1}
3
fX (x) = 14 , if x ∈ {2, 3}

0, otherwise


Show that Z = X 2 is a random variable. Find its p.m.f. and

distribution function.

I Solution:
Since h (x) = x 2 , x ∈ R, is a continuous function and X is a
random variable, it follows that Z = h (X ) = X 2 is a random
variable.

4 / 19
 Example 1 contd.

Clearly SX = {−2, −1, 0, 1, 2, 3} and SZ = {0, 1, 4, 9}.

Therefore, the pmf of Z is



 P(X = 0), if z = 0

P(X = −1) + P(X = 1), if z = 1



2
fZ (z) = P(X = z) = P(X = −2) + P(X = 2), if z = 4

P(X = −3) + P(X = 3), if z = 9





0, otherwise


5 / 19
 Example 1 contd.


1


 7, if z = 0
2
7, if z = 1



= 5
 14 , if z = 4
3
14 , if z = 9





0, otherwise


and the df of Z is


0, if z <0

1
7, if 0≤z <1



FZ (z) = 73 , if 1≤z <4.
 11
, if 4≤z <9


 14


1, if z ≥9


6 / 19
 Example 1

I In Example 1, directly find the df of Y = X 2 . Hence find the

p.m.f. of Y .
I Solution:
By Theorem 1, Y is a discrete r.v. with support
SY = {0, 1, 4, 9}. Thus, the df of Y is
G (z) = P(Y ≤ z) = P(X 2 ≤ z)


 0, if z <0

2 0≤z <1
P(X = 0), if



= P(X 2 = 0) + P(X 2 = 1), if 1≤z <4

P(X 2 = 0) + P(X 2 = 1) + P(X 2 = 4), if 4≤z <9





1, if z ≥9


7 / 19
 Example 1



 0, if z <0

1
7, if 0≤z <1



= 17 + 17 + 17 , if 1≤z <4
 1 1 1 1 3
7 + 7 + 7 + + 14 , if 4≤z <9



 7
1, if z ≥9



 0, if z <0

1
7, if 0≤z <1



= 37 , if 1≤z <4
 11
14 , if 4≤z <9





1, if z ≥9


8 / 19
 Example 1

The pmf of Y is
(
G (z) − G (z−), if z ∈ SY
g (z) =
0, o/w

1


 7, if z = 0
2
7, if z = 1



= 14 5
, if z = 4
 3
 ,

if z = 9


 14
0, otherwise


9 / 19
 Example 2

I Let X be a random variable with p.m.f.

(
|x|
, if x ∈ {±1, ±2, . . . , ±50}
fX (x) = 2550
0, otherwise

Show that Z = |X | is a random variable. Find its p.m.f., and

distribution function.

I Solution:
As h (x) = |x| , x ∈ R, is a continuous function and X is a
random variable, Z = |X | is a random variable.

We have SX = {± 1, ±2, · · · , ±50} and

SZ = {1, 2, · · · , 50}.

10 / 19
 Example 2 contd.
Therefore, the pmf of Z is
fZ (z) = P ({|X | = z}) = P ({X ∈ {−z, z}})
(
|−z|
+ |z| , if z ∈ {1, 2, . . . , 50}
= 2550 2550
0, otherwise
(
z
, if z ∈ {1, 2, . . . , 50}
= 1275
0, otherwise
and the df of Z is



 0, if z <1
 1 ,

if 1≤z <2
FZ (z) = 1275
i(i+1) .


 2550 , if i ≤ z < i + 1, i = 2, 3, . . . , 49

1, if z ≥ 50
11 / 19
 Example 3

I Let X be a random variable with p.m.f.

(

n x n−x
fX (x) = x p (1 − p) , if x ∈ {0, 1, . . . , n}
0, otherwise

where n is a positive integer and p ∈ (0, 1). Show that

Y = n − X is a random variable. Find its p.m.f. and
distribution function.

I Solution:
Note that SX = SY = {0, 1, · · · , n} and
h (x) = n − x, x ∈ R, is a continuous function. Therefore
Y = n − X is a random variable.

12 / 19
 Example 3 contd.

Thus the p.m.f. of Y is

fY (y ) = P ({X = n − y })
 !
n
p n−y (1 − p)y , if y ∈ {0, 1, . . . , n}


= n−y

0, otherwise

 !
 n p n−y (1 − p)y ,

if y ∈ {0, 1, . . . , n}
= y

0, otherwise


13 / 19
 Example 3 contd.

and the df of Y is

FY (y )


 0, if y < 0

n if 0 ≤ y < 1
p ,


 !
= Pi n
 j=0 (1 − p)j p n−j , if i ≤ y < i + 1, i = 1, 2, · · · , n − 1
j





1, if y ≥ n


14 / 19
 Function of a Continuous Random Variable
Theorem 3: Let X be a random variable of continuous type
with support SX and pdf fX (·).
Let S1 , S2 , . . . , Sk , be openSintervals in R such that
Si ∩ Sj = φ, if i 6= j and ki=1 Si = SX .
Let h : R → R be a Borel function such that, on each
Si (i = 1, . . . , k) , h : Si → R is strictly monotone and
continuously differentiable with inverse function hi−1 (·).
Let h (Sj ) = {h (x) : x ∈ Sj } so that h (Sj ) (j = 1, . . . , k) is
an open interval in R.
Then the rv T = h(X ) is a random variable of continuous
type with pdf
k
X   d
fT (t) = fX hj−1 (t) h−1 (t) Ih(Sj ) (t).
dt j
j=1

15 / 19
 Example 4

I Let X be random variable with p.d.f.

 |x|
2,
 if − 1 < x < 1
fX (x) = x3 , if 1 ≤ x < 2

0, otherwise


and let T = X 2 .

(i) Show that T is a random variable of absolutely continuous

type and hence, find the p.d.f. of T ;

(iii) Find the distribution function of T and hence find its p.d.f.;

16 / 19
 Example 4 contd.

I Solution:
(i) Clearly, T = X 2 is a random variable. We have
SX = (−1, 0) ∪ (0, 2) = S1 ∪ S2 , say.

h (x) = x 2 is strictly √
decreasing on S1 = (−1, 0) with inverse
function h−1 (t) = − t, t ∈ (0, 1), and h(S1 ) = (0, 1);

h (x) = x 2 is strictly√increasing on S2 = (0, 2) with inverse

function h−1 (t) = t, t ∈ (0, 4), and h(S2 ) = (0, 4).

17 / 19
 Example 4 contd.

I Using Theorem 3, it follows that T = X 2 is a random variable

of continuous type with p.d.f.

fT (t)
 √  d  √   √  d √ 
= fX − t − t I(0, 1) (t) + fX t t I(0, 4) (t)
dt dt
 √  1 √  1
= fX − t − √ I(0, 1) (t) + fX t √ I(0, 4) (t)
2 t 2 t
1 h  √  √  i
= √ fX − t I(0, 1) (t) + fX t I(0, 4) (t)
2 t

1
 2 , if 0 < t < 1

= 16 , if 1 < t < 4

0, otherwise


18 / 19
Thank you for listening!

19 / 19