Seoul National University Instructor: Junho Song

Dept. of Civil and Environmental Engineering [email protected]

457.212 Statistics for Civil & Environmental Engineers

In-Class Material: Class 09
Random Variable and Probability Distribution Functions (A&T: 3.1)

1. Random variable

(a) Definition: a variable quantity that takes on any value in a specified set according to
the assigned probabilities.

Probability

P(X=1) = 0.2
P(X=2) = 0.3
3 2
P(X=3) = 0.5
1 1
X= 3 3 2
x
Random variable Possible outcomes

(b) Discrete random variables

T: Number of tornadoes per year, { }

B: Number of substandard beams out of 10, { }

(c) Continuous random variables

C: Compact ratio of a soil specimen (0%: air only, 100%: no air), {c| }

2. Probability distribution functions

Discrete R.V. Continuous R.V

Probability or Density Probability Mass Function Probability Density Function

at X=x (PMF), PX(x) (PDF), fX(x)

Cumulative Frequency Cumulative Distribution Function

up to X=x (CDF), FX(x)

1
Seoul National University Instructor: Junho Song
Dept. of Civil and Environmental Engineering [email protected]

3. Probability Mass Function (PMF) of a discrete random variable X , PX (x)

(a) Definition

PX ( x) = P( )

(b) Example

X : Number of landfalls of hurricanes per year

x P(X=x)=
0 0.10
1 0.40
2 0.30
3 0.15
4 0.05

PX (2) =
PMF
PX (4) =
0.35
0.25
P(x)

0.15
0.05

0 1 2 3 4

2
Seoul National University Instructor: Junho Song
Dept. of Civil and Environmental Engineering [email protected]

# Creating PMF plot by R

x = 0:4
Px = c(0.1,0.4,0.3,0.15,0.05)
plot(x,Px,type="h",xlab="x",ylab="P(x)",main="PMF")

# Use plot() with "h" type rather than barplot() to have a correct scale
on x-axis
x = c(0,1.5,2,3,4)
plot(x,Px,type="h",xlab="x",ylab="P(x)",main="PMF")

(c) Properties

•  PX ( x) 

• P
all x
X ( x) = , e.g. PX (0) + ... + PX (4) =

• P ( a  X  b) = P X ( x) , e.g. P(0  X  2) =

4. Cumulative Distribution Function (CDF) of a discrete random variable X , FX (x)

(a) Definition:

FX ( x) = P( )

(b) Example

X : Number of landfalls of hurricanes per year

x PX(x) FX(x)=P(X≤x)
0 0.10
1 0.40
2 0.30
3 0.15
4 0.05

3
Seoul National University Instructor: Junho Song
Dept. of Civil and Environmental Engineering [email protected]

CDF

1.0
0.8
0.6
F(x)

0.4
0.2

0 1 2 3 4

# Creating CDF plot

Fx = cumsum(Px) # Creating CDF by cumulative sum of PMF
plot(x,Fx,type="s",xlab="x",ylab="F(x)",main="CDF")
lines(x,Fx,type="p",pch=16)

(c) Properties

• FX ( a ) = P X ( x)

• FX (−) =

because P(X − ) =

• FX () =

because P( X ) =

• 𝑃(𝑎 < 𝑋 ≤ 𝑏) = −

e.g. P(0  X  2) =

4
Seoul National University Instructor: Junho Song
Dept. of Civil and Environmental Engineering [email protected]

Example 1: A company owns two buildings, A and B. The probabilities that the buildings A
and B will be non-operational after an earthquake are 0.2 and 0.1, respectively. The
buildings are located closely to each other. As a result, the probability that the building B is
non-operational is twice the original probability if the building A is non-operational.

(a) What is the probability that there is no building in operation?

(b) What is the probability that both buildings are in operation?

(c) Let X denote the number of operational buildings. Find and plot the probability mass
function (PMF) of X .

(d) Find and plot the cumulative distribution function (CDF) of X .

x PX(x) FX(x)=P(X≤x)

5
Seoul National University Instructor: Junho Song
Dept. of Civil and Environmental Engineering [email protected]

5. Probability Density Function (PDF) of a continuous random variable X , f X (x)

(a) Definition: “Density” of probability distribution at X = x.

N→∞
Bin size → 0

Normalized Polygon Probability Density Function

P( x  X  x + x) = f X ( x)  x

f X ( x) = lim

(b) Properties

• f X (x) 0


• f
−
X ( x)dx = ; “Marginalization rule”

• P ( a  X  b) = 

6
Seoul National University Instructor: Junho Song
Dept. of Civil and Environmental Engineering [email protected]

6. Cumulative Distribution Function (CDF) of a continuous random variable X , FX (x)

(a) Definition

FX ( x ) = P ( )

=  f X ( x)dx

f X (x) FX (x)

x x

(b) Properties

dFX ( x)
• = . In summary, [PDF] [CDF]
dx

• Non-decreasing (because we are integrating a non-negative function)

• FX (−) = P( )=

• FX () = P( )=

FX (x)