Chapter 2.

The Random Variable

0. Introduction
1. The Random Variable Concept
2. Distribution Function
3. Density Function
4. The Gaussian Variable
5. Other Distribution and Density Examples
6. Conditional Distribution and Density Functions

Chapter 2. The Random Variable 1

2.1 Random Variable Concept

 Random variable: Real function of the

elements of a sample space S.
 RV is represented
p byy a capital
p letter
such as W,X, or Y and any particular
value of the RV by a lowercase letter
such as w
w, xx, or yy.
 Given an experiment defined by a
sample space S with elements s, we
assign to every s a real number X (s)
according to some rule
 X : a function that maps all elements
of the sample space into points on
the real line or some part thereof

Chapter 2. The Random Variable 2

2.1 Random Variable Concept

‐ Example 2.1‐1
2.1 1
 Rolling a die and flipping a coin
 Sample space
 Let the r.v. be a function X as follows
1. A coin head (H) outcome
‐> positive value shown up on the die
2. A coin tail (T) outcome
‐> negative and twice value shown up on the die

Chapter 2. The Random Variable 3

2.1 Random Variable Concept

‐ Example 2.1‐2
2.1 2

 The pointer on a wheel of chance is spun

 The possible outcomes are the numbers from 0
to 12 marked on the wheel
 Sample
p space
p {{0 < s  12}}
 Define a r.v. by the function
X  X (s)  s 2

 Points in S map onto the real line as the set {0

< x  144}

Chapter 2. The Random Variable 4

2.1 Random Variable Concept
 Conditions for a Function to be a RV
 Every point in S must correspond to only one
value of the RV
 The set { X  x } shall be an event for any real
number
b x
The probability of this event, P{X  x}, is equal to the sum
of the probabilities of all the elementary events
corresponding to {X  x}
 The probabilities of the events {X=} and {X=‐ }
be 0: P{ X  }  0 P{ X  }  0
 Discrete and Continuous RVs
 Discrete RV: Having only discrete
Example 2.1‐1 : discrete RV defined on a discrete sample
space
From Example 2.1‐2 : we could define a discrete RV as
having the value 1 for the set of outcomes {0 < s  6},
and ‐1 for {6 < s  12}. This discrete RV is defined on a
continuous sample space

 Continuous RV : Having a continuous range of

values
Example 2.1‐2 : continuous RV defined on a continuous
sample
l space
 Mixed r.v. : Some of its values are discrete and
some are continuous
Chapter 2. The Random Variable 5
2. 2 Distribution Function
 Cumulative Probabilityy Distribution
Function (CDF) F X ( x )  P { X  x }
 The probability of the event {X  x}
 Depend on x
 A function of x
 We shall often call FX(x) just the
distribution function of X.
 Find the CDF graph for Rolling a Die

 Example of CDF

Chapter 2. The Random Variable 6

2. 2 Distribution Function
Properties of CDF
(1 ) F X (  ) 
(2) F X ( ) 
(3) ?  FX ( x)  ?
(4) if x1  x 2 ,
(5 ) P { x1  X  x 2 } 
(6 ) F X ( x  or  )  F X ( x )

 The
h fourth
f h states that
h FX(x)
( ) is
i a non decreasing
d i ft.
f
of x
 The fifth property is justified from the fact that
h events X  x1 andd x1  X  x2  are
the
mutually exclusive so the probability of the event
X  x2   X  x1 x1  X  x2  is the sum fo the
probabilities PX  x1 andd Px1  X  x2 
 Properties 1,2,4, and 6 ay be used as tests to
determine if some function could be a valid
distribution function.

Chapter 2. The Random Variable 7

2. 2 Distribution Function

 Discrete Random Variable

 If X is discrete RV, FX (x) : stairstep form
 A
Amplitude
li d off a step : probability
b bili off occurrence off the
h
value where the step occurs
 If the values of X are denoted xi, we may write
N
FX (x)  P{X  xi }u(x  xi )
i 1 1 x  0
where u(∙) is the unit‐step function u ( x)  
0 x  0
 Using P( xi )  P{ X  xi }
N N
FX ( x)   P{ X  xi }u ( x  xi )   P( xi )u ( x  xi )
i 1 i 1

Chapter 2. The Random Variable 8

2. 2 Distribution Function
– Example 2.2
2.2‐1
1
 Let X have the discrete values in the set {‐1, ‐0.5,
0.7, 1.5, 3}
 The corresponding probabilities {0.1, 0.2, 0.1, 0.4,
0.2}

Chapter 2. The Random Variable 9

2. 2 Distribution Function
– Example 2.2
2.2‐2
2
 Wheel‐of‐chance experiment

Chapter 2. The Random Variable 10

2.3 Density Function
 Density function of RV, X dFX ( x)
f X ( x) 
dx
 Existence
 If the derivative of FX(x) exists,
exists then fX(x) exists.
exists
 However, there may be places where dFX(x)/dx
is not defined
eg Points of abrupt change in slope
eg.
 Continuous RV
Have Continuous distribution FX(x)
FX(x) also have corners as in example 2.2‐2
fX(x) : a function with step‐type discontinuities

 Discrete RV
Stairstep form distribution function
Unit‐impulse function, (t) is used for describing the
derivative of FX(x) at the stairstep points.

N
FX ( x)   P( xi )u ( x  xi )
i 1
N
f X ( x)   P( xi ) ( x  xi )
i 1
Chapter 2. The Random Variable 11
2.3 Density Function

 Unit‐Impulse Function (t)

 Definition by its integral property

 ( x0 )    ( x) ( x  x0 )dx
• (x) : any continuous function at the point x = x0
• (t) ; Can be integrated as a “function” with infinite
amplitude, area of unity, and zero duration
• The relationship of unit‐impulse and unit‐step functions
du ( x) x
 ( x) 
d
dx
or   ( )d  u ( x)
• The general impulse function
Shown symbolically
y y as a vertical arrow occurringg at
the point x=x0 and having an amplitude equal to
the amplitude of the step function for which it is
the derivative

[Link]

Chapter 2. The Random Variable 12

2.3 Density Function

 A discrete
di t RV
 The density function
N
for a discrete RV FX ( x)   P( xi )u ( x  xi )
exists i 1
N
 Using impulse  f X ( x)   P( xi ) ( x  xi )
functions to describe i 1
the
h d derivative off FX(x)
( )
at its stair step points

 Properties of Density
Functions
(1)  f X ( x) all x

( 2) 
f X ( x)dx 
(3) FX ( x) 
(4) P{x1  X  x2 } 
Chapter 2. The Random Variable 13
2.3 Density Function

– Example 2.3‐1
2.3 1
 Test the function gX(x) if it can be a valid
density function
 Property 1 : nonnegative
 Property 2: Its area a=1 ‐> a=1/

0 x0    x  x 0  

 1
g X ( x )   2 ( x  x0   ) x0    x  x0

1 1
  ( x  x0 ) x 0  x  x0  
   2

0 x0    x
 x
 1
x0 
g X ( ) d   ( x  x 0   ) 2
x0    x  x0
2 2
G X ( x)  
 1  x g ( )d  1  1 ( x  x )  1 ( x  x ) 2 x  x  x  
 2 x0 X 2 
0
2 2
0 0 0

1 x  x0  

Chapter 2. The Random Variable 14

2.3 Density Function

– Example 2.3‐2,
2.3 2, 2.3
2.3‐3
3
 Example 2.3‐2
 From example 2.3‐1, set x0= 8,  = 5, and a = 1/  = 1/5.
 Find the probability that X has values greater than 4.5
but not greater than 6.7.

 
0 x0    x  x0   0 3  x  13
 
 1  ( x  3)
f X ( x )   2 ( x  x0   ) x0    x  x 0   3 x 8
  25
1 1  ( x  8)
  ( x  x0 ) x 0  x  x0   0 . 2  8  x  13
  2  25
6.7
6.7 x  3 1  x2 
P{4.5  X  6.7}   dx    3x  0.2288
4.5 25 25  2  4.5

  
x2
 Example 2.3‐3 FX ( x)  u ( x) 1  e b  where b  0
 
 Find its density function  

 
x2   
x2 
dF ( x)  du ( x) d
f X ( x)  X  1 e b   u ( x) 1  e b 
dx   dx dx  
   
 
x2 

x2

x2
  2x 2x
 1  e b  ( x)  u ( x) e b  u ( x) e b
  b b
 

Chapter 2. The Random Variable 15

2.4 Gaussian RV
 A RV,, X is called Gaussian if its densityy
function has the form
( xax )2

1 2 X2
f X ( x)  e
2 X2
(  a x ) 2 No closed-

1 x 2 X2
FX ( x )   e d form solution
2 X2

where  X  0 and
   aX  

Chapter 2. The Random Variable 16

2.4 Gaussian RV

 Numerical or approximation methods for

Gaussian r.v.
 Tables ‐> many tables according to variable  X , a X
 Only
O l one table
t bl according
di to li d  X  1, a X  0
t normalized
 Consider
(  a x ) 2

1 x 2 X2
FX ( x )   e d
2 X2
2
1 x 
 F(X ) 
2
 e 2 d

 For a negative value of x ;

 From FX(x), consider
F ( x)  1  F ( x)

  aX d
u  du  ,    X u  aX
X X

2
1   a X  u2
   xa X 
1 x 2  X 1
FX ( x ) 
 X 2
 e  d 
2
 X e 2 du

 x  aX 
 FX ( x)  F  
 X 
Chapter 2. The Random Variable 17
2.4 Gaussian RV

– Example 2.4‐1,
2.4 1, 2.4
2.4‐2
2

 Example 2.4‐1
 Find the probability of the event {X5.5} for Gaussian r.v.
having aX = 3 and X = 2
P{ X  5.5}  FX (5.5)
 x  aX    a X 5.5  3
FX ( x)  F   u   1.25
 X  X 2
P{ X  5.5}  FX (5.5)  F (1.25)  0.8944

 Example 2.4‐2
 Assume that the height of clouds at some location is
Gaussian r.v. X with aX = 1830m and X = 460m
 Find the probability that clouds will be higher than
2750m
P{ X  2750}  1  P{ X  2750}
F ( x)  1  F ( x)  1  FX (2750)
 2750  1830 
 1 F 
 460 
 1  F (2.0)
 1.0  0.9772
 0.0228

Chapter 2. The Random Variable 18

2.4 Gaussian RV

 Evaluations of F(X) by approximation

2
1  
F ( x)  1  Q( x) where Q( x) 
2
x e 2 d
 2
1  
Q( x) 
2
x e 2 d Qf
Q-function
i

x2

 1  e 2
  x0
 (1  a ) x  a x 2  b  2
  a  0.339, b  5.510

 Ex) 2.4‐3 Gaussian r.v. with aX = 7, X = 0.5

 7.3  7 
P{ X  7.3}  FX (7.3)  F    F (0.6)  0.7257
 0 .5 
F (0.6)  1  Q(0.6)
   ( 0. 6 ) 2 / 2
 1   e
1
 0.661(0.6)  0.339 (0.6) 2  5.51  2
 
 0.7264
Chapter 2. The Random Variable 19
2.5 Other CDF/PDF Examples

 Binomial 0  p  1 and N  1,2, 

 Density function & distribution function
N
N k
f X ( x)     p (1  p) N  k ( x  k )
k 0  k 
N
N k
FX ( x)     p (1  p) N  k u ( x  k )
k 0  k 

 For N=6
and p=0.25

Chapter 2. The Random Variable 20

2.5 Other CDF/PDF Examples

 Poisson b  0 (real constant )

 Density function & distribution function

bk
f X ( x )  e  ( x  k )
b

k  0 k!

bk
FX ( x)  e  u ( x  k )
b

k 0 k!

 Quite similar to those for the binomial r.v.

 If N and p0 for the binomial case in such a way
that Np=b,
Np=b a constant,
constant the Poisson case results
 Applications: Wide variety of counting‐type applications
1. The number of defective units in sample taken from a
production line
2. The number of telephone calls made during a period of
time
3. The number of electrons emitted from a small section of
cathode in a given time interval
 If the time interval of interest has duration T, and the
events being counted are known to occur at an average
rate  and have a Poisson distribution, then b = T.

Chapter 2. The Random Variable 21

2.5 Other CDF/PDF Examples

 Uniform    a   and b  a
 1  0 xa
 a xb xa
f X ( x)   (b  a ) FX ( x )   a xb
 0 elsewhere b  a
 1 bx
 The
quantization of
signal samples
prior to
encoding
di ini
digital
communication
systems
‐> The error
introduced in
the round‐off
process
‐> uniform
distributed

Chapter 2. The Random Variable 22

2.5 Other CDF/PDF Examples

 Exponential    a   and b  a

 1  ( x b a )  1  ( x b a )
 xa 
f X ( x)   b e FX ( x)  1  b e xa
 0 xa  0 xa

 Describing
rainfrop sizes
when a large
number of
rainstorm
measurements
are made
 Fluctuations in
signal strength
received
i d by
b
radar from
certain tyoes of
air craft

Chapter 2. The Random Variable 23

2.5 Other CDF/PDF Examples

 Rayleigh    a   and b  a
2 ( xa ) 2

 ( x  a )e b xa
f X ( x)   b
 0 xa

 ( xa ) 2


FX ( x)  1  e xa
b

 0 xa

 Y=(X12+X22)1/2
X1 and X2 are
zero‐mean
statistically
independent
Gaussian RVs
( 0)
(a=0).
 The envelope of
one type of
noise
i when
h
passed through
a bandpass filter
Chapter 2. The Random Variable 24
2.6 Conditional Distribution and Density
Functions

 Conditional Probability
 For two events A and B where P(B)( )0,, the
conditional probability of A given B
P( A  B)
P( A | B) 
P (B
( B)
 Conditional Distribution
 A : Identified as the event {X  x} for the r.v. x
 Conditional distribution function of X
P{ X  x  B}
P{ X  x | B}   FX ( x | B)
P( B)
 { X  x  B} = the joint event { X  x}  B
This joint event consists of all outcomes s such that
X ( s )  x and sB

 The conditional distribution : discrete,

continuous, or mixed random variables

Chapter 2. The Random Variable 25

2.6 Conditional Distribution and Density
Functions

 Properties of Conditional Distribution

(1) FX ( | B) 
(2) FX ( | B ) 
(3)  FX ( x | B) 
(4)  if x1  x2
(5) P{x1  X  x2 | B}

(6) FX ( x  /  | B)  FX ( x | B )

 These characteristics have the same general

meanings as described earlier following eqs.
(2.2‐2) : Distribution function section

Chapter 2. The Random Variable 26

2.6 Conditional Distribution and Density
Functions

 Conditional Density
 Conditional density function of the r.v. x
The derivative of the conditional distribution
function
dFX ( x | B)
f X ( x | B) 
dx

If FX(x|B) contains discontinuities, impulse

response are present in fX(x|B) to account for
the derivatives at the discontinuities
 Properties of Conditional Density
(1) 0  f X ( x | B) all x

( 2) 
f X ( x | B)dx  1
x
(3) FX ( x | B )  
f X ( | B)d
x2
(4) P{x1  X  x2 | B}  x1
f X ( x | B)dx

Chapter 2. The Random Variable 27

2.6 Conditional Distribution and Density
Functions

 Method of Defining Conditioning Event

 Event B is defined in terms of the RV X
 Event B may depend on some RV other than X
 Example: B  { X  b}   b  , b is real no.

P{ X  x  B}
P{ X  x | B}   FX ( x | B)
P( B)
P{ X  x  X  b}
 FX ( x | X  b)  P{ X  x | X  b} 
P ( X  b)
1. bx인 경우 ( X  x )  ( X  b)  ( X  b)
P{ X  x  X  b}
FX ( x | X  b)  P{ X  x | X  b} 
P ( X  b)
P ( X  b)
 1
P ( X  b)
2. x<b인 경우
( X  x )  ( X  b)  ( X  x )

P{ X  x  X  b}
FX ( x | X  b)  P{ X  x | X  b} 
P ( X  b)
P( X  x)

P ( X  b)

Chapter 2. The Random Variable 28

2.6 Conditional Distribution and Density
Functions

 Method of Defining Conditioning Event(con’d)

 FX ( x)
 xb
FX ( x | X  b)   FX (b)
 1 xb
 f X ( x) f X ( x)
  b
xb
f X ( x | X  b)   FX (b)  f X ( x)dx

 xb
 0

• Conditional Event has

not zero Prob.
0  FX (b)  1

 FX ( x | X  b)  FX ( x)
 f X ( x | X  b)  f X ( x ) x  b

Chapter 2. The Random Variable 29

Problems
 2.1‐7, 9, 10, 11, 12, 13, 14
 2.2‐5, 9, 10
 2.3‐1, 2, 3, 4, 5, 6, 7, 8, 15
 2.4‐1, 2, 3, 4, 5, 6, 7, 10, 11, 12, 14
 2.5
2 5‐1
1, 3,
3 4,
4 5,
5 6,
6 7
 2.6‐1, 2, 3

Chapter 2. The Random Variable 30