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

4-Random Variables

Uploaded by

dinhlinhna0510
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
12 views

4-Random Variables

Uploaded by

dinhlinhna0510
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 80

Random Variables

1 / 76
Objectives
1 Understand random variables (RVs) which
are classified into two types: discrete and
continous RVs
2 Basic concept of distribution: cumulative
distribution; probability mass function for
discrete RVs; probability density function
for continuous RVs and their connections
2 / 76
Intro
• Sample space and events are basic
components of probability
• Similar to numbers in calculus
• Study the relations between numbers we
use functions
• In probability we use random variables
3 / 76
Table of contents
1 Definition of Random Variables
2 Events Defined by Random Variable
3 Cumulative Distribution Function
4 Discrete Random Variables
5 Continuous Random Variable
4 / 76
Definition
A random variable X defined on a sample space
Ω is a quantity that is calculated by the outcomes
- a function of outcomes

5 / 76
Example

• Toss a coin three times


• X be the number of times that tails appear
• Sample space Ω =
{HHH, HHT, HTH, HTT, THH, THT, TTH, TTT}
• X(HHH) = 0, X(HHT) = 1 . . .

6 / 76
w HHH HHT HTH HTT THH THT TTH TTT
X(w) 0 1 1 2 1 2 2 3

All possible values of X are 0, 1, 2, 3. We say Range(X) =


{0, 1, 2, 3}
1
P(X = 0) = P(there is no tail) = P(HHH) =
8
3
P(X = 1) = P(there is 1 tail) = P({HHT, HTH, THH}) =
8

7 / 76
Example
• Tossing a fair coin until a Head appear.
The sample space is
Ω = {H, TH, TTH, TTTH, . . . }
• X: the number of tossing
• Range(X) = {1, 2, 3, . . . ...}
• X = 1 if and only if the first coin turns head
so P(X = 1) = P(first coin turns H) = 0.5
8 / 76
Table of contents
1 Definition of Random Variables
2 Events Defined by Random Variable
3 Cumulative Distribution Function
4 Discrete Random Variables
5 Continuous Random Variable
9 / 76
Event defined by Random variable
{X = a} = {w ∈ Ω : X(w) = a}
{X ≥ a} = {w ∈ Ω : X(w) ≥ a}
{X < a} = {w ∈ Ω : X(w) < a}
{a ≤ X ≤ b} = {w ∈ Ω : a ≤ X(w) ≤ b}
{a ≤ X < b} = {w ∈ Ω : a ≤ X(w) < b}
{a < X ≤ b} = {w ∈ Ω : a < X(w) ≤ b}
{a < X < b} = {w ∈ Ω : a < X(w) < b}
10 / 76
Probability associated with a
random variable X

P(X = a), P(X ≥ a), P(X > a),


P(X ≤ b), P(X < b),
P(a ≤ X ≤ b), P(a < X ≤ b),
P(a ≤ X < b), P(a < X < b)
11 / 76
Table of contents
1 Definition of Random Variables
2 Events Defined by Random Variable
3 Cumulative Distribution Function
4 Discrete Random Variables
5 Continuous Random Variable
12 / 76
Cumulative distribution function
(c.d.f)
cdf F(.) of the random variable X is a function
defined by

F(b) = P(X ≤ b)

is the probability that X takes on values less than


or equal to b
13 / 76
Properties of c.d.f

1 F(x) is a nondecreasing function in x


2 0 ≤ F(x) ≤ 1
3 lim F(x) = 0, lim F(x) = 1
x→−∞ x→∞

14 / 76
Use c.d.f to answer questions about a random variable

P(X ≤ b) = F(b)
P(X < b) = lim P(X ≤ b − h) = lim F(x) = F(b−)
h→0+ x→b−
P(X > a) = 1 − P(X ≤ a) = 1 − F(a)
P(a < X ≤ b) = P(X ≤ b) − P(X ≤ a) = F(b) − F(a)
P(a ≤ X ≤ b) = F(b) − F(a−)
P(a < X < b) = F(b−) − F(a)
15 / 76
Example
The c.d.f of a random variable X is defined by

0
 if x < 0
F(x) = x + 0.5 if 0 ≤ x < 0.5

1 if x ≥ 0.5

Then
P(X > 0.25) = 1 − F(0.25) = 1 − (0.25 + 0.5) = 0
16 / 76
Example
X is a random variable with c.d.f
(
0 if x < 1
F(x) =
k 1 − e−(x−1)

if x ≥ 1

1 Find the value of k


2 Compute F(3), P(X > 2) and
P(2 < X ≤ 3)
17 / 76
Solution
1 Using the property lim F(x) = 1, we have
x→∞
 
lim k 1 − e−(x−1) = 1 ⇒ k = 1
x→∞
2

F(3) = 1 − e−2
P(X > 2) = 1 − F(2) = 1 − e−1
P(2 < X ≤ 3) = F(3) − F(2) = e−1 − e−218 / 76
Exercise
The c.d.f of the time T it takes a bank teller to
serve a customer is defined by

0
 if x < 2
F(x) = A(x − 2) if 2 ≤ x < 6

1 if x ≥ 6

1 Find the value of the constant A


2 P(X > 4) =? P(3 ≤ X ≤ 5) =? 19 / 76
Types of RV
Based on range of the random variable X
• If the set of possible values of X is finite or
countable like {0, 1, 2, 3}, {1, 2, . . . } then
X is called discrete RV
• If the set of possible values of X is
uncountable (like the interval [a, b],
[a, ∞)) then X is called continuous RV
20 / 76
Table of contents
1 Definition of Random Variables
2 Events Defined by Random Variable
3 Cumulative Distribution Function
4 Discrete Random Variables
5 Continuous Random Variable
21 / 76
Probability mass function
• A discrete random variable is a random
variable that can take on at most a
countable number of possible values.
• The probability mass function (p.m.f) of
the discrete random variable X is defined as

p(xi) = P(X = xi) for all xi ∈ Range(X)


22 / 76
Example
Roll two fair dice then the sample space is

Ω = {(1, 1, ), . . . , (6, 6)} = {(i, j) : 1 ≤ i, j ≤ 6}

Let X be the largest of numbers on two dice, i.e


if the rolling result is (i, j) then

X(i, j) = max(i, j)
23 / 76
Table values of X

24 / 76
All possible values of X is

Range(X) = {1, 2, 3, 4, 5, 6}

so X is a discrete random variable.


In order to determine the pmf of X, we need to
find all the probabilities

P(X = 1), P(X = 2), . . . , P(X = 6)

25 / 76
X = 1 if and only if the outcome is (1, 1). So

P(X = 1) = P((1, 1)) = 1/36

X = 2 if and only if the outcomes is one of


(1, 2), (2, 2), (2, 2). So

P(X = 2) = P({(1, 2), (2, 1), (2, 2)})


= P((1, 2)) + P((2, 1)) + P((2, 2) = 3/36

26 / 76
Similar, we have P(X = 3) = 5/36, P(X = 4) =
7/36, P(X = 5) = 9/36, P(X = 6) = 11/36.

We can summary p.m.f of X in the p.m.f table


x 1 2 3 4 5 6
1 3 5 7 9 11
P(X = x) 36 36 36 36 36 36

27 / 76
Similar, we have P(X = 3) = 5/36, P(X = 4) =
7/36, P(X = 5) = 9/36, P(X = 6) = 11/36.

We can summary p.m.f of X in the p.m.f table


x 1 2 3 4 5 6
1 3 5 7 9 11
P(X = x) 36 36 36 36 36 36

27 / 76
Illustration to calculate p.m.f

For each possible value x, we collect all the outcomes that


give rise to X = x and add their probabilities to obtain
pX (x) = P(X = x). 28 / 76
One can use p.m.f of the discrete random vari-
able X to answer any question of X such as
P(1 < X < 4) = P(X = 2 or X = 3)
= P(X = 2) + P(X = 3) = 3/36 + 5/36
using additive rule for disjoin set
P(A ∪ B) = P(A) + P(B) if A ∩ B = ∅
for A = {X = 2}, B = {X = 3}
29 / 76
Practice
Probability mass function of discrete random vari-
able X is
x -2 -1 0 1 2
p(x) = P(X = x) 1/8 2/8 2/8 2/8 1/8
Determine
1 P(X ≤ −1 or X = 2)

2 P(−1 ≤ X ≤ 1)
30 / 76
Practice

A shipment of 20 similar laptop computers to a


retail outlet contains 3 that are defective. If a
school makes a random purchase of 2 of these
computers, find the probability mass function
(p.m.f) for the number of defectives.

31 / 76
Practice

An urn contains 11 balls, 3 white, 3 red, and 5


blue balls. Take out 2 balls at random, without
replacement. You win $1 for each red ball you
select and lose a $1 for each white ball you se-
lect. Determine the p.m.f of your loss/profit X.

32 / 76
Properties of p.m.f
• X is discrete
→ Range(X) = {x1, . . . , xn . . . }
• p(xi) = P(X = Xxi) ≥ 0
• P(X ∈ A) = p(xi)
xi ∈A
• Normalization

X
P(−∞ < X < ∞) = 1 ⇒ p(xi) = 1
i=1 33 / 76
Properties of p.m.f
• X is discrete
→ Range(X) = {x1, . . . , xn . . . }
• p(xi) = P(X = Xxi) ≥ 0
• P(X ∈ A) = p(xi)
xi ∈A
• Normalization

X
P(−∞ < X < ∞) = 1 ⇒ p(xi) = 1
i=1 33 / 76
Properties of p.m.f
• X is discrete
→ Range(X) = {x1, . . . , xn . . . }
• p(xi) = P(X = Xxi) ≥ 0
• P(X ∈ A) = p(xi)
xi ∈A
• Normalization

X
P(−∞ < X < ∞) = 1 ⇒ p(xi) = 1
i=1 33 / 76
Example
Suppose X has 3 values 1, 2, 3 and
1 1
p(1) = , p(2) =
2 3
then what is p(3)?

p(3) = 1 − p(1) − p(2) = 1/6.


34 / 76
Example
Suppose X has 3 values 1, 2, 3 and
1 1
p(1) = , p(2) =
2 3
then what is p(3)?

p(3) = 1 − p(1) − p(2) = 1/6.


34 / 76
Graph of p(x)

35 / 76
Practice
Suppose that the p.m.f of random variable X is
given by

p(x) = c(x + 5), x = 0, 1, 2, 3, 4

Find
1 the value of the constant c
2 P(0 < X < 2.5).
36 / 76
From pdf to cdf

Probability that X does not exceed a given value


X
F(b) = P(X ≤ b) = P(X = xi)
xi ≤b

37 / 76
Example
Suppose that p.m.f of X is given by p(1) = 21 , p(2) =
1 1
3 , p(3) = 6 then

F(0.5) = P(X ≤ 0.5) = 0


5
F(2.4) = P(X ≤ 2.4) = P(X = 1)+P(X = 2) =
6
38 / 76
The formula of the c.d.f of X is



 0, x < 1
1, 1 ≤ x < 2

F(x) = 25
6, 2 ≤ x < 3



1, x ≥ 3

39 / 76
Graph of F(x)

Remark
Jump size at 1 is P(X = 1), ... 40 / 76
From c.d.f to p.m.f

If X is a discrete random variable with c.d.f F(x)


then the p.m.f of X is determined by

P(X = x) = P(X ≤ x)−P(X < x) = F(x)−F(x−)

where x is a jump discontinuity of F

41 / 76
Example
Determine the p.m.f of X from the c.d.f



 0 if x < −2

0.2 if − 2 ≤ x < 0
F(x) =


 0.7 if 0 ≤ x < 2
if x ≥ 2

1

42 / 76
Solution
Graph of F(x) is

P(X = a) = F(a)−F(a−) is nonzero at the points −2, 0, 2.


The p.m.f at each point is the change (jump size) of c.d.f
at the point 43 / 76
p.m.f of X is given by
p(−2) = P(X = −2) = F(−2) − F(−2−)
= 0.2 − 0 = 0.2
p(0) = P(X = 0) = F(0) − F(0−)
= 0.7 − 0.2 = 0.5
p(2) = P(X = 2) = F(2) − F(2−)
= 1 − 0.7 = 0.3

44 / 76
Properties of cdf of a discrete RV

• lim F(x) = 0 and lim F(x) = 1


x→−∞ x→∞
• FX has a piecewise constant and
staircase-like form.

45 / 76
Practice
c.d.f of discrete random variable X is given by

0
 if x < 1
F(x) = 0.7 if 1 ≤ x < 3

1 if x ≥ 3

Compute
1 P(X ≤ 2) and P(X > 2)

2 P(1 ≤ X ≤ 2)
46 / 76
Table of contents
1 Definition of Random Variables
2 Events Defined by Random Variable
3 Cumulative Distribution Function
4 Discrete Random Variables
5 Continuous Random Variable
47 / 76
RVs with Uncountable Values
• If Range(X) is countable, we can list all
possible value of random variable X
between a and b to evaluate
X
P(a ≤ X ≤ b) = P(X = xi)
a≤xi ≤b
• If X is a continuous random variable, i.e.
Range(X) is uncountable, it leads to
”uncountable sum” 48 / 76
Probability density functions of
continuous R.V
Suppose Range(X) is uncountable.
X is continuous if there is a non negative function f (x) so
that
Z b
P(a ≤ X ≤ b) = f (x) dx
a
|{z}
probability density function (p.d.f) of X

49 / 76
Interpretation of p.d.f
Z a+ϵ
P (a ≤ X ≤ a + ϵ) = f (x)dx ≈ ϵf (a)
a

f (a) is a measure of how likely it is that the


random variable will be near a - ”pmf per unit
length”
50 / 76
f (x)dx ≈ P(x < X < x + dx) ≈ P(X ≈ x)
So
X X
P(a ≤ X ≤ b) ≈ P(X ≈ x) = f (x)dx
a≤X≤b a≤x≤b
Z b
= f (x)dx
a
51 / 76
Example
Suppose that the error in the reaction tempera-
ture, in ◦C, for a controlled laboratory experi-
ment is a continuous random variable X having
the probability density function
(2
x
−1 ≤ x ≤ 2
f (x) = 3 .
0 elsewhere
52 / 76
We have
f (0.3) = f (−0.3) = 0.03 < f (0.6) = 0.12
So the error in the reaction temperature near 0.3

C is as likely as near -0.3◦C but has less chance
than near 0.12 ◦C
f (3) = 0
implies that the error can not be near 3 ◦C.
53 / 76
f (x) = 0 ∀x ≥ 2 or x ≤ −1
says that the error can not be greater than 2. It
can not also take on values less than -1. The
error can be between -1 and 2 or we say

Range(X) = [−1, 2]

which is uncountable
54 / 76
Use pdf to answer questions of continuous random variable

For example
1 1
x2
Z Z
P(0 ≤ X ≤ 1) = f (x)dx = dx =
0 0 3

55 / 76
Z 1
P(X ≤ 1) = P(−∞ < X ≤ 1) = f (x)dx
−∞
Z −1 Z 1
= f (x)dx + f (x)dx
Z−∞
−1 Z −1
1 2 Z 1 2
x x
= 0dx + dx = dx
−∞ −1 3 −1 3

56 / 76
Remark that Range(X) = [−1, 2] then X ≤ 1 is
equivalent to −1 ≤ X ≤ 1. So we have
Z 1 2
x
P(X ≤ 1) = P(−1 ≤ X ≤ 1) = dx
−1 3

57 / 76
Properties of continuous RV
• P(X = a) = 0 for all a
• For all a, b

P(a ≤ X ≤ b) = P(a < X ≤ b)


= P(a ≤ X < b)
= P(a < X < b)

58 / 76
Probability as an Area

59 / 76
Exercise
Suppose that X has p.d.f
(
3
256 (8x − x2) if 0 < x < 8
f (x) =
0 elsewhere

Determine
(a) P(X = 3) (b) P(2 < X < 4)
(c) P(X > 6) (d) P(X < 5)
60 / 76
Conditions of pdf
• f (x) ≥ 0 for all x
• (Normalization) P(−∞ < X < ∞) = 1 implies
that
Z∞
f (x)dx = 1
−∞

61 / 76
Example
A gambler spins a wheel of fortune, continuously cali-
brated between 0 and 1, and observes the resulting num-
ber. Assuming that all subintervals of [0, 1] of the same
length are equally likely. The observed number is a ran-
dom variable X with pdf
(
c, 0 ≤ x ≤ 1
f (x) =
0, otherwise

62 / 76
The constant c is determined by
Z ∞
f (x)dx = 1
−∞

So Z 1
cdx = 1
0
then c = 1
63 / 76
Practice
Suppose the p.d.f of X is
n C(4x − 2x2) 0 < x < 2
f (x) =
0 otherwise
What is the value of the constant C?
Find P(X > 1)?
64 / 76
From p.d.f to c.d.f

Let X be a continous random variable with p.d.f


f (x) then its c.d. f is given by
Z x
F(x) = P(X ≤ x) = f (u)du
−∞

65 / 76
Properties of cdf of a continuous RV

• F ′(x) = f (x) for all x


• P(a ≤ X ≤ b) = F(b) − F(a)
• lim F(x) = 0 and lim F(x) = 1
x→−∞ x→∞

66 / 76
Example
Suppose that the error in the reaction temperature, in ◦C,
for a controlled laboratory experiment is a continuous ran-
dom variable X having the probability density function
(2
x
−1 < x < 2
f (x) = 3 .
0 elsewhere

Find c.d.f F of X
67 / 76
Solution
• For x < −1
Z x Z x
F(x) = f (u)du = 0du = 0
−∞ −∞

• For −1 ≤ x ≤ 2
Z x x
u2 x3 + 1
Z
F(x) = f (u)du = du =
−∞ −1 3 9

68 / 76
• For x > 2
x 2
u2
Z Z
F(x) = f (u)du = du = 1
−∞ −1 3

• Hence the c.d.f of X is



0,
 x < −1
3
F(x) = x 9+1 , −1 ≤ x ≤ 2

1, x>2

69 / 76
Exercise
The lifetime (in hours) of an electric component
is a continous random variable X with p.d. f
(
0.1e−0.1x if x ≥ 0
f (x) =
0 otherwise

1 Find the c.d.f F(x) of X.


2 What is the probability that the component
will last at least 5 hours? 70 / 76
From c.d.f to p.d.f

If F is the c.d.f of a continous random variable


X then the p.d.f of X is
d
f (x) = F(x)
dx

71 / 76
Example

Find p.d.f of X if its c.d.f is


(
1 − e−2x if x > 0
F(x) =
0 elsewhere

72 / 76
Solution

(
′ 2e−2x if x > 0
f (x) = F (x) =
0 elsewhere

73 / 76
content...

74 / 76
content...

75 / 76
content...

76 / 76

You might also like