0% found this document useful (0 votes)
62 views

Module 14 Expectation of A Random Variable

The document provides definitions and results about the expectation of a random variable. It defines the expectation of a discrete and absolutely continuous random variable as a weighted sum or integral of the possible values. It then proves that the expectation can also be defined using probability distribution functions. The document also shows that if a function is applied to a random variable, the expectation of the result is equal to applying the function to the expectation of the variable. It provides examples to illustrate the concepts and results.

Uploaded by

pankaj kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
62 views

Module 14 Expectation of A Random Variable

The document provides definitions and results about the expectation of a random variable. It defines the expectation of a discrete and absolutely continuous random variable as a weighted sum or integral of the possible values. It then proves that the expectation can also be defined using probability distribution functions. The document also shows that if a function is applied to a random variable, the expectation of the result is equal to applying the function to the expectation of the variable. It provides examples to illustrate the concepts and results.

Uploaded by

pankaj kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 20

Module 14

Expectation of a Random Variable

() Module 14 Expectation of a Random Variable 1 / 19


X : a given r.v. with d.f. FX (·) and p.m.f./p.d.f. fX (·).
Definition 1:
(a) Let X be a discrete r.v. with support SX and p.m.f. fX (·).
We say that the expected value of X (denoted by E (X ))
exists and equals
X
E (X ) = x fX (x),
x∈SX
P
provided x∈SX |x|fX (x) < ∞.
(b) Let X be an A.C. r.v. with p.d.f. fX (·). We say that the
expected value of X (denoted by E (X )) exists and equals
Z ∞
E (X ) = x fX (x)dx,
−∞
R∞
provided −∞ |x|fX (x)dx < ∞.

() Module 14 Expectation of a Random Variable 1 / 19


Result 1:
Let X be a discrete or A.C. r.v. Then
Z ∞ Z 0
E (X ) = P({X > y })dy − P({X < y })dy ,
0 −∞

provided the expectation exists.


Proof: For A.C. case (the proof for discrete case follows similarly).
Z ∞ Z 0 Z ∞
E (X ) = x fX (x)dx = x fX (x)dx + x fX (x)dx
−∞ −∞ 0
Z 0 Z 0 Z ∞Z x
= − fX (x)dydx + fX (x)dydx
−∞ x 0 0
Z 0 Z y Z ∞Z ∞
= − fX (x)dxdy + fX (x)dxdy
−∞ −∞ 0 y
Z 0 Z 0
= − P({X < y })dy + P({X > y })dy .
−∞ ∞

() Module 14 Expectation of a Random Variable 2 / 19


Corollary 1:

(a) Let X be a discrete or an A.C. r.v. with P({X ≥ 0}) = 1.


Then
Z ∞ Z ∞
E (X ) = P({X > y }) = (1 − FX (y ))dy .
0 0

(b) If X is discrete with P({X ∈ {0, 1, 2, ....}}) = 1, then



X
E (X ) = P({X ≥ k}).
k=1

() Module 14 Expectation of a Random Variable 3 / 19


Result 2:
Let X be a discrete or an A.C. r.v. with p.m.f./p.d.f. fX (·) and support
SX . Let Y = g (X ), for some function g : R → R. Then
 P
 x∈SX g (x)fX (x), if X is discrete
E (Y ) = E (g (X )) = .
 R∞
−∞ g (x)fX (x)dx, if X is A.C.
Proof: (For discrete case.) Let fY (·) be the p.m.f. of Y and let
SY = g (SX ) = {g (x) : x ∈ SX }. Then, clearly, SY is the support of Y and
X
E (Y ) = y fY (y )
y ∈SY
X
= y P({h(X ) = y })
y ∈SY
X X
= y fX (x)
y ∈SY x∈SX
h(x)=y

() Module 14 Expectation of a Random Variable 4 / 19


X X
= y fX (x)
y ∈SY x∈SX
h(x)=y
X X
= h(x)fX (x)
y ∈SY x∈SX
h(x)=y
X
= h(x)fX (x).
x∈SX

() Module 14 Expectation of a Random Variable 5 / 19


Example 1:

Let X be a r.v. with p.m.f.


 cp
 xp , if x = 1, 2, . . .
fX (x) = ,
0, otherwise

where p > 1 is a given constant. Then SX = {1, 2, . . .} and


X
E (|X |r ) = |x|r fX (x)
x∈SX

X 1
= cp
x p−r
x=1

is finite iff r < p − 1. Thus E (X r ) exists iff r < p − 1.

() Module 14 Expectation of a Random Variable 6 / 19


In particular E(X) exists iff p > 2. For p > 2
X
E (X ) = x fX (x)
x∈SX

X 1
= cp .
x p−1
x=1

() Module 14 Expectation of a Random Variable 7 / 19


Example 2:
Let X be a r.v. with p.d.f.
1 1
fX (x) = · , −∞ < x < ∞.
π 1 + x2
Then
Z ∞
E (|X |) = |x|fX (x)dx
−∞

|x|
Z
1
= dx
π −∞ 1 + x2
Z ∞
2 x
= dx
π 0 1 + x2

= ∞.

Thus E (X ) does not exists.


() Module 14 Expectation of a Random Variable 8 / 19
Example 3:

Let X be a r.v. with p.d.f.



 2x, if 0 < x < 1
fX (x) = .
0, otherwise

Then
Z ∞
E (X 3 ) = x 3 fX (x)dx
−∞
Z 1
= 2 x 4 dx
0

2
= .
5

() Module 14 Expectation of a Random Variable 9 / 19


Result 3:

Let X be a discrete or an A.C. r.v. and let gi : R → R, i = 1, ..., k, be


such that E (gi (X )) exists.
(a) If P({g1 (X ) ≤ g2 (X )}) = 1 then E (g1 (X )) ≤ E (g2 (X )). In
particular if P({a ≤ X ≤ b}) = 1, for some real constants a
and b, then a ≤ E (X ) ≤ b.
(b) If P({X ≥ 0}) = 1 and E (X ) = 0 then P({X = 0}) = 1.
(c) If E (X ) exists then |E (X )| ≤ E (|X |).
(d) For real constants c1 , . . . , ck
k k
!
X X
E ci gi (X ) = ci E (gi (X )) .
i=1 i=1

() Module 14 Expectation of a Random Variable 10 / 19


Proof :
(a) (For discrete case.) Let A = {x ∈ R : g1 (x) ≤ g2 (x)}. Then,
fX (x) = 0, ∀ x ∈ Ac (P({g1 (X ) ≤ g2 (X )}) = 1).
X
E (g1 (X )) = g1 (x)fX (x)
x∈SX
X X
= g1 (x)fX (x) + g1 (x)fX (x)
x∈SX ∩A x∈SX ∩Ac
X
= g1 (x)fX (x)
x∈SX ∩A
X
≤ g2 (x)fX (x)
x∈SX ∩A
X X
= g2 (x)fX (x) + g2 (x)fX (x)
x∈SX ∩A x∈SX ∩Ac
X
= g2 (x)fX (x) = E (g2 (X )).
x∈SX
() Module 14 Expectation of a Random Variable 11 / 19
(b) (Discrete case.) Since P({X ≥ 0}) = 1, we have
SX ⊆ [0, ∞). Thus, for n = 1, 2, . . .,
X
E (X ) = x fX (x)
x∈SX
X X
= x fX (x) + x fX (x)
x∈SX ∩[0, n1 ) x∈SX ∩[ n1 ,∞)
X
≥ x fX (x)
x∈SX ∩[ n1 ,∞)
1 X
≥ fX (x)
n
x∈SX ∩[ n1 ,∞)
 
1 1
= P X ≥
n n
 
1
⇒ P X ≥ ≤ nE (X ) = 0, ∀ n = 1, 2, . . .
n
() Module 14 Expectation of a Random Variable 12 / 19
 
1
⇒ P X ≥ = 0, ∀ n = 1, 2, . . .
n
 
1
⇒ lim P X ≥ = 0
n→∞ n
∞  !
[ 1
⇒ P X ≥ = 0
n
n=1

⇒ P({X > 0}) = 0

⇒ P({X = 0}) = 1.

() Module 14 Expectation of a Random Variable 13 / 19


(c) Follow from (a) on using the fact that

−|X | ≤ X ≤ |X |.

(d) (Discrete case.) For simplicity of notations we prove the


result for k = 2.
X
E (c1 g1 (X ) + c2 g2 (X )) = (c1 g1 (x) + c2 g2 (x))fX (x)
x∈SX
X
= c1 g1 (x)fX (x) +
x∈SX
X
c2 g2 (x)fX (x)
x∈SX
= c1 E (g1 (X )) + c2 E (g2 (X )).

Remark 1: If E (X ) exists then using (c) above it follows that


|E (X )| < ∞ (i.e., E (X ) is finite).
() Module 14 Expectation of a Random Variable 14 / 19
Some special Expectations

X a r.v.;
g : R → R: a given function;
Then Y = g (X ) is a r.v. and E (g (X )) = expected value of g (X ).
Some special expectations are:
(i) µ01 = µ = E (X ) = mean of (distribution of) X;
(ii) For r ∈ {1, 2, ...} , µ0r = E (X r ) = r -th moment of X about origin;
(iii) For r ∈ {1, 2, ....} , E (|X |r ) = r -th absolute moment of X about
origin;
(iv) For r ∈ {1, 2, ...} , µr = E ((X − µ)r ) = r -th moment of X about its
mean (or r -th central moment);
(v) µ2 = σ 2 = E ((X − µ)2 ) = Variance of X (written as Var(x)).

() Module 14 Expectation of a Random Variable 15 / 19


Remark 1:

(a)

Var(X ) = E ((X − µ)2 )


= E (X 2 − 2µX + µ2 )
= E (X 2 ) − 2µE (X ) + µ2
= E (X 2 ) − µ2
= E (X 2 ) − (E (X ))2 .

(b) Since Var(X) = E ((X − E (X ))2 ) ≥ 0, we have

E (X 2 ) ≥ (E (X ))2 ,

for any r.v. X .


(c) Var(X ) = 0 ⇒ P({X = E (X )}) = 1.

() Module 14 Expectation of a Random Variable 16 / 19


Take Home Problems

1 Let X be a r.v. with p.d.f.



c (x + 1), if − 1 < x < 1
fX (x) = ,
0, otherwise

where c is a real constant.


(a) Find the value of c;
(b) Find the mean and variance of X .
2 . Let X be a r.v. with p.d.f.
1

 x2
, if x > 1
fX (x) = .
0, otherwise

Show that E (X ) does not exists.

() Module 14 Expectation of a Random Variable 17 / 19


Abstract of Next Module

In next module we will introduce a transform


 
MX (t) = E e tX , t ∈ R,

that can be used to generate moments (µ0r = E (X r ), r = 1, 2, . . .) of a


r.v. X . This transform is called the moment generating function (m.g.f.)
of X . We will study various properties of m.g.f.

() Module 14 Expectation of a Random Variable 18 / 19


Thank you for your patience

() Module 14 Expectation of a Random Variable 19 / 19

You might also like