Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

MSA 8190 - Fall 2023

Statistical Foundation
Continuous Probability Distributions

Mohammad Javad Feizollahi

September 11, 2023

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 1 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Lecture’s Objectives

1 Understand the concept and nature of different continuous

probability distributions such as:
Uniform
Triangular
Normal
Exponential, Erlang, Gamma
Weibull
Chi-square, t, F
2 Compute pdf, cdf, expected value, and variance for
continuous random variables
3 Perform previous objective in R

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 2 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Continuous Distributions

When the variable being measured is expressed on a

continuous scale, its probability distribution is called a
continuous distribution.
Examples:
1 Diameter of piston rings,
2 Viscosity of oil,
3 Customers’ waiting time.
Probability density function (pdf) f (x):
1 f (x) ≥ 0, for all x ∈ R
R +∞
2
−∞
f (x)dx = 1
Rb
3 P(a ≤ X ≤ b) = a
f (x)dx for any a, b ∈ R s.t. a ≤ b
4 P(X = a) = 0 ̸= f (a)

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 3 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Probability Density Function (pdf)

Z b
P(a ≤ X ≤ b) = f (x)dx
a

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 4 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

CDF, Mean and Variance of a Continuous RV

Z x
F (x) = P(X ≤ x) = f (u)du
−∞
Z +∞
µ = E(X ) = xf (x)dx
−∞
  Z +∞ h i2
2 2
σ = E (X − µ) = (x − µ)2 f (x)dx = E(X 2 ) − E(X )
−∞

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 5 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Continuous Uniform Random Variable

1
A continuous RV X with pdf f (x) = , a ≤ x ≤ b is a
b−a
continuous uniform random variable.

(b − a)2
µ = E(X ) = and σ 2 = Var(X ) =
12


 0 x <a
F (x) = a≤x <b

1 b≤x


© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 6 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Example with Triangular Distribution

The product quality characteristic measurement is a random

variable X with probability distribution
(
4(x − 11.75) 11.75 ≤ x ≤ 12.25
f (x) =
4(12.75 − x) 12.25 ≤ x ≤ 12.75

1 If the quality specification limit is that X must be less than

12.50, what is the percentage of products to be
nonconforming?
2 Compute F (x), E(X ) and Var(X ).

Solution:

P(X ≥ 12.5) =

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 7 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

PDF Curve

2.0
1.5
f(x)

1.0
0.5
0.0

11.5 12.0 12.5 13.0

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 8 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Define and Plot the PDF

1 > f<- function(x){

2 + y<-x
3 + pts<- length(x)
4 + for(i in 1:pts){
5 + ifelse( x[i]<11.75, y[i]<-0,
6 + ifelse( x[i] <= 12.25, y[i]<-(4*(x[i]-11.75)),
7 + ifelse(x[i] <= 12.75, y[i]<-(4*(12.75-x[i])),
8 + y[i]<-0)))
9 + }
10 + return(y)
11 + }
12 > curve(f, from=11.5, to=13,n=1001)

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 9 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Probability of an Interval
Z 12.75
P(X ≥ 12.5) = f (x)dx = 0.125
12.5
2.0
1.5
f(x)

1.0
0.5
0.0

11.5 12.0 12.5 13.0

x
© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 10 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Shade an Area Under Plot

1 > curve(f, from=11.5, to=13,n=1001)

2 > x <- seq(12.5,12.75,0.001)
3 > y1 <- 0*x
4 > y2 <- f(x)
5 > polygon(c(x,rev(x)),c(y2,rev(y1)),col="skyblue")

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 11 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Numerical Integration to Compute P(·) E(X ) and Var(X )

1 > integrate(f, lower = 11.75, upper = 12.75)

2 1 with absolute error < 1.1e-14
3 > integrate(function(x) x*f(x), lower = 11.75,
4 + upper = 12.75)
5 12.25 with absolute error < 1.4e-13
6 > integrate(function(x) (x-12.25)^2*f(x),
7 + lower = 11.75,upper = 12.75)
8 0.04166667 with absolute error < 4.6e-16

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 12 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

CDF
Z x
F (x) = f (u)du
−∞
0.0 0.2 0.4 0.6 0.8 1.0
F(x)

11.5 12.0 12.5 13.0

x
© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 13 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

R Code to Compute and Plot CDF

1 > F<-function(x){
2 + y<-x
3 + pts<- length(x)
4 + for(i in 1:pts){
5 + ifelse( x[i]<11.75, y[i]<-0,
6 + ifelse( x[i] <= 12.75,
7 + y[i]<-( integrate(f, lower = 11.75,
8 + upper = x[i])$val), y[i]<-1))
9 + }
10 + return(y)
11 + }
12 > curve(F, from=11.5, to=13,n=1001)

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 14 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Normal Distribution

A RV X with pdf

1 −(x−µ)2
f (x) = √ e 2σ2 , x ∈ R
2πσ
is a normal random variable with parameters µ and σ, where
µ ∈ R, and σ > 0.
E(X ) = µ and Var(X ) = σ 2
X ∼ N(µ, σ 2 ) means X has a normal pdf with parameters µ
and σ.

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 15 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Normal PDF

f (x)

! – 3! µ – 2! !–! ! ! + ! ! + 2! ! + 3! x

68%
95%
99.7%
© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 16 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Standard Normal Distribution

N(0, 1) is the standard normal pdf. A RV with N(0, 1) is

denoted by Z .
If X is a normal RV with E(X ) = µ and Var(X ) = σ 2 , then
x −µ
Z= is a standard normal RV and
σ
 
X −µ x −µ
P(X ≤ x) = P ≤ = P(Z ≤ z)
σ σ

x −µ
where z = is the z-value obtained by standardizing X .
σ

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 17 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Example

The diameter of a shaft in an optical storage drive is normally

distributed with mean 0.2508 inch and standard deviation 0.0005
inch. The specifications on the shaft are 0.2500 ± 0.0015 inch.
What proportion of shafts conforms to specifications?
Solution:
Let X denote the shaft diameter in inches. Then
P(0.2485 < X < 0.2515) =
= 0.91924

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 18 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Example

f (x) Specifications

0.2485 0.2508 0.2515 x

0.25

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 19 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Another Example

Three shafts are made and assembled in a linkage. The length of

each shaft, in centimeters, is distributed as follows:
Shaft 1: N (75, 0.09)
Shaft 2: N (60, 0.16)
Shaft 3: N (25, 0.25)
Assume the shafts’ length are independent to each other:
1 What is the distribution of the linkage?
2 What is the probability that the linkage will be longer than
160.5 cm?

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 20 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Solution
Let Y1 ∼ N(µ1 , σ12 ), Y2 ∼ N(µ2 , σ22 ) and Y3 ∼ N(µ2 , σ22 )
denote normal RVs corresponding to the length of shafts 1, 2,
and 3, respectively.
Let X denotes the RV for the linkage length.
Then X ∼ N(µ, σ 2 ) where,
µ = µ1 + µ2 + µ3 = 75 + 60 + 25 = 160 and
σ 2 = σ12 + σ22 + σ32 = 0.09 + 0.16 + 0.25 = 0.5. Therefore,
X ∼ N(160, 0.5). √
P(X > 160.5) = P(Z > 160.5−160
√
0.5
) = P(Z > 0.5) = 0.23975
Solution in R

1 > pnorm(160.5,160,sqrt(0.5),lower.tail = FALSE)

2 [1] 0.2397501
3 > pnorm(sqrt(0.5),lower.tail = FALSE)
4 [1] 0.2397501

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 21 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Normal Approximation to the Binomial Distribution

X − np
If X is a binomial RV, then Z = p is approximately a
np(1 − p)
standard normal RV. The approximation is good for np > 5 and
n(1 − p) > 5.
0.25

n = 10
p = 0.5

0.20

0.15
f(x)

0.10

0.05

0.00

©
0 1 2 3 4 5 6 7 8 9 10
x
2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 22 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Example

In a digital communication channel, assume that the number of

bits received in error can be modeled by a binomial RV, and
assume that the probability that a bit is received in error is 10−5 .
If 16 million bits are transmitted, what is the probability that more
than 150 errors occur?

Check assumptions: np = 160 >> 5, n(1 − p) = 15999840 >> 5

150
X
P(X > 150) = 1 − Cx16,000,000 (10−5 )x (1 − 10−5 )16,000,000−x
x=0
= pbinom(150, 1.6 ∗ 1e7, 1e − 5, lower .tail = F ) = 0.77197
150 − 160
≈ P(Z > √ ) = P(Z > −0.7906) = 0.7854
159.998

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 23 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Continuity Correction for Normal Approximations

We can use a continuity correction to improve the quality of this

approximation by computing
150 − 160+0.5
P(Z > √ ) = P(Z > −0.7510) = 0.7737.
159.998

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 24 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Normal Approximation to the Hypergeometric Distribution

Binomial distribution is a satisfactory approximation to the

hypergeometric distribution when k, the sample size, is small
relative to m + n, the size of the population from which the
k
sample is selected ( m+n < 0.1).
Therefore, the normal distribution can provide an effective
approximation of hypergeometric probabilities when
k m
m+n < 0.1, kp > 5 and k(1 − p) > 5, where p = m+n .

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 25 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Normal Approximation to the Poisson Distribution

If X is a Poisson RV with E(X ) = λ, then Z = X√−λ λ
is
approximately a standard normal RV. The approximation is
good for λ > 5.
Example: Assume that the number of asbestos particles in a
squared meter of dust on a surface follows a Poisson
distribution with a mean of 1000. If a squared meter of dust is
analyzed, what is the probability that less than 950 particles
are found?

λ = 1000
950 −1000
X e 1000x
P(X ≤ 950) =
x!
x=0
950 − 1000
≈ P(Z ≤ √ ) = P(Z ≤ −1.5811) = 0.057
1000
© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 26 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Linear Combination of Random Variables

Given RVs X1 , X2 , · · · , Xn and constants c1 , c2 , · · · , cn ∈ R,

Y = c1 X1 + c2 X2 + · · · + cn Xn

is a linear combination of X1 , X2 , · · · , Xn .

E(Y ) = c1 E(X1 ) + c2 E(X2 ) + · · · + cn E(Xn )

Var(Y ) = c12 Var(X1 ) + c22 Var(X2 ) + · · · + cn2 Var(Xn )

Pn Pn
+2 ci cj Cov(Xi , Xj )
i=1 j=i+1

If X1 , X2 , · · · , Xn are independent

Var(Y ) = c12 Var(X1 ) + c22 Var(X2 ) + · · · + cn2 Var(Xn )

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 27 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Mean and Variance of a Sample Average

Consider an independent and identically distributed (iid)

sample of n RVs X1 , X2 , · · · , Xn .
Suppose E(Xi ) = µ and Var(Xi ) = σ 2 , ∀i = 1, · · · , n.
Pn
Xi
X̄ = i=1 is called the sample average.
n
Then, X̄ is itself a RV with

E(X̄ ) = and Var(X̄ ) =

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 28 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Reproductive Property of the Normal Distribution

Suppose X1 , X2 , · · · , Xn are independent, normal RVs with

E(Xi ) = µi and Var(Xi ) = σi2 , for i = 1, · · · , n.
Then, Y = c1 X1 + c2 X2 + · · · + cn Xn is a normal random
variable
with

E(Y ) =
and
Var(Y ) =

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 29 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Exponential Distribution

The RV X that equals the distance between successive counts of a

Poisson process with mean λ > 0 is an exponential RV with
parameter λ. The pdf of X is f (x) = λe −λx for x ∈ R+ .

1
F (x) = , µ = E(X ) = and σ 2 = Var(X ) =
λ2

Lack of Memory Property: P(X > x + y |X > y ) = P(X > x)

Proof:

An exponential RV is the continuous analog of a

, and they share a similar lack of memory
property.
The exponential distribution is often used in reliability studies
as the model for the time until failure of a device.
© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 30 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Shape of Exponential Distribution

2.0

λ
2
0.5
1.6 0.1

1.2

f (x)

0.8

0.4

0.0
0 2 4 6 8 10 12
x
© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 31 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Erlang Distribution

The RV X that equals the interval length until r counts occur in a

Poisson process with mean λ > 0 is an Erlang RV with parameters
r r −1 e −λx
λ and r . The pdf of X is f (x) = λ x(r −1)! for x ∈ R++ and
r = 1, 2, · · · .

F(x) is computed via integration by parts

µ = E(X ) = and σ 2 = Var(X ) =
An Erlang RV can be thought of as the continuous analog of
a random variable.

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 32 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Shape of Erlang Distribution

2.0

r λ
1 1
1.6 5 1
5 2

1.2

f (x)

0.8

0.4

0.0

0 2 4 6 8 10 12
x

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 33 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Gamma Distribution
The Erlang distribution is a special case of the gamma
distribution. If the parameter r of an Erlang RV is not an
integer, but r > 0, the RV has a gamma distribution.
Need a generalization of the factorial function
Gamma function:
Z ∞
Γ(r ) = x r −1 e −x dx, for r > 0
0
Γ(r ) = (r − 1)Γ(r − 1) for r ∈ R++
Γ(r ) = (r − 1)! for r ∈ N

λr x r −1 e −λx
Gamma RV X has pdf f (x) = with parameters λ > 0
Γ(r )
and r > 0. For integer r , X has an Erlang distribution.

r r
and σ 2 = Var(X ) = 2
© µ = E(X ) =
2023 M. J. Feizollahi, All Rights Reserved λ λ
Continuous Probability Distributions 34 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Shape of Gamma Distribution

2.0

r λ
1 1
1.6 8.3 2
7.5 3.75

1.2

f (x)

0.8

0.4

0.0

0 2 4 6 8 10 12
x

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 35 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Weibull Distribution

The Weibull distribution is used to model the time until failure of

many different physical systems in which the number of failures
increases with time (bearing wear),
decreases with time (some semiconductors),
or remains constant with time (failures caused by external
shocks to the system).

Weibull RV X with scale parameter δ > 0 and shape parameter

β > 0 has pdf:
β  x β−1 −( x )β
f (x) = e δ
δ δ
An additional parameter of θ location is included in the translated
Weibull distribution

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 36 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Weibull Distribution

For a Weibull RV X with scale parameter δ > 0 and shape

parameter β > 0:
 β
x
− 
δ
F (x) = 1 − e
1
)
µ = E(X ) = δΓ(1 +
β
1 2
 
2 2 2 2
σ = Var(X ) = δ Γ(1 + ) − δ Γ(1 + )
β β

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 37 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Shape of Weibull Distribution

1.0

δ β
1 1
0.8 3.4 2
4.5 6.2

0.6

f (x)

0.4

0.2

0.0

0 3 6 9 12 15
x

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 38 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Lognormal Distribution

Suppose W ∼ N(θ, ω 2 ). Then, X = e W is a lognormal RV with

pdf
(ln x − θ)2
1 −
f (X ) = √ e 2ω 2 , for x ∈ R+
xω 2π

2 /2 2 2
E(X ) = e θ+ω and Var(V ) = e 2θ+ω (e ω − 1)
The lifetime of a product that degrades over time is often modeled
by a lognormal RV.

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 39 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Chi-square Distribution

A Chi-square RV X with ν ∈ N degrees of freedom has pdf

x ν
e (− 2 ) x ( 2 −1)
f (x) = ν/2 .
2 Γ(ν/2)

λr x r −1 e −λx
Note that a gamma RV X has pdf f (x) = with
Γ(r )
parameters λ > 0 and r > 0.
Then, chi-square distribution is a special case of gamma
distribution with parameters λ = 1/2 and r = ν/2.
E(X ) = and Var(X ) = .
Which Chi-square distribution is equivalent to an exponential
distribution?

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 40 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Construction of Chi-square Distributions

If Z1 , · · · , Zk are independent, standard normal RVs, then

k
X
X = Zi2 ∼ χ2 (k)
i=1

If Xi ∼ χ2 (νi ) for i = 1, · · · , n, and they are independent,

then
Xn
Y = Xi ∼
i=1

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 41 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Shape of Chi-square Distribution with k DoF

f (x)

k=2

k=5

k = 10

0 5 10 15 20 25 x

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 42 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Student’s t Distribution

A RV X has t distribution with ν ∈ N degrees of freedom if its pdf

has the following form:

Γ ν+1


2
f (x) =
√
 x2 ( ν+1 ) , x ∈ R
ν 2
πνΓ 2 ν +1

ν
E(X ) = 0 and Var(X ) = , for ν > 2
ν−2

If Z ∼ N(0, 1) and V ∼ χ2 (ν) are independent RVs, then

Z
p ∼ t(ν)
V /ν
t distribution tends to standard normal distribution as ν
increases (ν > 30)
© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 43 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Shape of t Distribution

Comparison of t Distributions

0.4
Distributions
df=1
df=3
df=8
df=30
0.3

normal
0.2
f(x)

0.1
0.0

-4 -2 0 2 4

x value

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 44 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

F Distribution

Let W and Y be independent Chi-square RVs with u and v

degrees of freedom, respectively. Then the ratio

W /u
F =
Y /v

has the F distribution with u degrees of freedom in the numerator

and v degrees of freedom in the denominator which is
u+v u

u
x (u/2)−1


Γ 2 v
2
f (x) =  u+v , x ≥ 0
Γ u2 Γ v2


 u

v x + 1 2

v 2v 2 (v + v − 2)
E(X ) = , for v > 2 and Var(X ) = , for v > 2
v −2 u(v − 2)2 (v − 4)
© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 45 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Shape of F Distribution

f (x)

u = 5, v = 5

u = 5, v = 15

0 2 4 6 8 10 x

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 46 / 48

Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Useful Facts About Continuous Distributions

1 A gamma distribution with shape parameter r > 0 and rate
parameter λ > 0 is an Erlang distribution if r = 1, 2, 3, · · · .
2 It is an Exponential distribution if r = 1.
3 Sum of k independent gamma RVs with the same λ, is itself a
gamma RV.
4 A gamma RV with r = ν/2 and λ = 1/2 is χ2 (ν).
5 If W is a Weibull RV with scale δ and shape β, then

β
X = Wδ has an exponential distribution with rate δ.
6 2
χ (2) is equivalent to the exponential distribution with
λ = 1/2.
7 If Z ∼ N(0, 1) and V ∼ χ2 (ν) are independent RVs, then
Z
p ∼ t(ν)
V /ν
8 If X ∼ t(ν), then X 2 ∼ F (1, ν)
© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 47 / 48
Intro Uniform Normal Exponential Erlang Gamma Weibull Lognormal Chi-square t F

Continuous Distributions in R

Distribution cdf in R
Beta pbeta
Cauchy pcauchy
Chi-Square pchisq
Exponential pexp
F pf
Gamma pgamma
Lognormal plnorm
Normal pnorm
Pareto ppareto (Package: actuar)
t pt
Uniform punif
Weibull pweibull

© 2023 M. J. Feizollahi, All Rights Reserved Continuous Probability Distributions 48 / 48