Beruflich Dokumente
Kultur Dokumente
net/publication/319306526
CITATIONS READS
0 255
1 author:
Americo Cunha Jr
Rio de Janeiro State University
119 PUBLICATIONS 181 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Flight Control and Trajectory Planning of Balloon-Multicopter Aerial Vehicles View project
All content following this page was uploaded by Americo Cunha Jr on 27 August 2017.
americo@ime.uerj.br
www.americocunha.org
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 1 / 60
Probability in Dimension 1
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 2 / 60
Random experiment
An experiment which repeated under same fixed conditions produce
differente results is called random experiment.
Examples:
1 Rolling a cube-shaped fare die
3 Measuring temperature
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 3 / 60
Probability space
Σ: σ-algebra on Ω
(set with relevant events only)
P: probability measure
(measure of expectation of an event occurrence)
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 4 / 60
Sample space
Examples:
1 Rolling a cube-shaped fare die (finite Ω)
Ω = {1, 2, 3, 4, 5, 6}
2 Choosing an integer even number randomly (denumerable Ω)
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 5 / 60
σ-algebra of events
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 6 / 60
Examples of σ-algebras
Are σ-algebras:
Σ = φ, {1, 3, 5}, {2, 4, 6}, Ω
Σ = φ, {1, 2, 3, 4}, {5, 6}, Ω
Σ = 2Ω (set of all subsets)
G. Grimmett and D. Welsh, Probability: An Introduction. Oxford University Press, 2 edition, 2014.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 7 / 60
Probability measure
Remark:
P {φ} = 0 (empty set has probability zero)
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 8 / 60
An example in discrete probability
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 9 / 60
An example in discrete probability
Probability space 1:
Ω1 = {(H,H), (H,T ), (T ,H), (T ,T )}
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 9 / 60
An example in discrete probability
Probability space 1:
Ω1 = {(H,H), (H,T ), (T ,H), (T ,T )}
P1 {(H,H)} = 1/4, P1 {(H,T )} = 1/4,
P1 {(T ,H)} = 1/4, P1 {(T ,T )} = 1/4
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 9 / 60
An example in discrete probability
Probability space 1:
Ω1 = {(H,H), (H,T ), (T ,H), (T ,T )}
P1 {(H,H)} = 1/4, P1 {(H,T )} = 1/4,
P1 {(T ,H)} = 1/4, P1 {(T ,T )} = 1/4
Probability space 2:
Ω2 = {0, 1, 2}
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 9 / 60
An example in discrete probability
Probability space 1:
Ω1 = {(H,H), (H,T ), (T ,H), (T ,T )}
P1 {(H,H)} = 1/4, P1 {(H,T )} = 1/4,
P1 {(T ,H)} = 1/4, P1 {(T ,T )} = 1/4
Probability space 2:
Ω2 = {0, 1, 2}
P2 {0} = 1/4, P2 {1} = 1/2, P2 {2} = 1/4
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 9 / 60
An example in continuous probability
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 10 / 60
An example in continuous probability
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 10 / 60
An example in continuous probability
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 10 / 60
An example in continuous probability
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 10 / 60
Remarks on probability
Probability zero
An impossible event has probability zero
(e.g. roll a six faces dice, numbered from 1 to 6, and get 7)
Not every event with probability zero is impossible
(e.g. randomly pick a point on the main diagonal of a square)
Probability one
An event whose occurrence is certain has probability one
(e.g. throw a coin and obtain head or tail)
Not every event with probability one occurs
(e.g. randomly pick a point outside square’s main diagonal)
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 11 / 60
Conditional probability
It follows that
P {A ∩ B} = P A | B × P {B} .
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 12 / 60
An example in conditional probability
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 13 / 60
An example in conditional probability
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 13 / 60
An example in conditional probability
P {A} = 6/36
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 13 / 60
An example in conditional probability
P {A} = 6/36
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 13 / 60
An example in conditional probability
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 13 / 60
An example in conditional probability
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 13 / 60
An example in conditional probability
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 13 / 60
Independence of events
P {A ∩ B} = P {A} × P {B} .
Remark:
This notion generalizes itself naturally to n events.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 14 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
Probability space 1 (fair deck):
2♦,3♦,4♦,5♦,6♦,7♦,8♦,9♦,10♦,J♦,Q♦,K ♦,A♦,
2♣,3♣,4♣,5♣,6♣,7♣,8♣,9♣,10♣,J♣,Q♣,K ♣,A♣,
cards =
2♥,3♥,4♥,5♥,6♥,7♥,8♥,9♥,10♥,J♥,Q♥,K ♥,A♥,
2♠,3♠,4♠,5♠,6♠,7♠,8♠,9♠,10♠,J♠,Q♠,K ♠,A♠
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
Probability space 1 (fair deck):
2♦,3♦,4♦,5♦,6♦,7♦,8♦,9♦,10♦,J♦,Q♦,K ♦,A♦,
2♣,3♣,4♣,5♣,6♣,7♣,8♣,9♣,10♣,J♣,Q♣,K ♣,A♣,
cards =
2♥,3♥,4♥,5♥,6♥,7♥,8♥,9♥,10♥,J♥,Q♥,K ♥,A♥,
2♠,3♠,4♠,5♠,6♠,7♠,8♠,9♠,10♠,J♠,Q♠,K ♠,A♠
P1 {Q } = 4/52
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
Probability space 1 (fair deck):
2♦,3♦,4♦,5♦,6♦,7♦,8♦,9♦,10♦,J♦,Q♦,K ♦,A♦,
2♣,3♣,4♣,5♣,6♣,7♣,8♣,9♣,10♣,J♣,Q♣,K ♣,A♣,
cards =
2♥,3♥,4♥,5♥,6♥,7♥,8♥,9♥,10♥,J♥,Q♥,K ♥,A♥,
2♠,3♠,4♠,5♠,6♠,7♠,8♠,9♠,10♠,J♠,Q♠,K ♠,A♠
P1 {Q } = 4/52
P1 {♠} = 13/52
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
Probability space 1 (fair deck):
2♦,3♦,4♦,5♦,6♦,7♦,8♦,9♦,10♦,J♦,Q♦,K ♦,A♦,
2♣,3♣,4♣,5♣,6♣,7♣,8♣,9♣,10♣,J♣,Q♣,K ♣,A♣,
cards =
2♥,3♥,4♥,5♥,6♥,7♥,8♥,9♥,10♥,J♥,Q♥,K ♥,A♥,
2♠,3♠,4♠,5♠,6♠,7♠,8♠,9♠,10♠,J♠,Q♠,K ♠,A♠
P1 {Q } = 4/52
P1 {♠} = 13/52
P1 {Q♠} = 1/52
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
Probability space 1 (fair deck):
2♦,3♦,4♦,5♦,6♦,7♦,8♦,9♦,10♦,J♦,Q♦,K ♦,A♦,
2♣,3♣,4♣,5♣,6♣,7♣,8♣,9♣,10♣,J♣,Q♣,K ♣,A♣,
cards =
2♥,3♥,4♥,5♥,6♥,7♥,8♥,9♥,10♥,J♥,Q♥,K ♥,A♥,
2♠,3♠,4♠,5♠,6♠,7♠,8♠,9♠,10♠,J♠,Q♠,K ♠,A♠
P1 {Q } = 4/52
P1 {♠} = 13/52
P1 {Q♠} = 1/52 = 4/52 × 13/52
|{z} | {z }
P1 {Q} P1 {♠}
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
Probability space 1 (fair deck):
2♦,3♦,4♦,5♦,6♦,7♦,8♦,9♦,10♦,J♦,Q♦,K ♦,A♦,
2♣,3♣,4♣,5♣,6♣,7♣,8♣,9♣,10♣,J♣,Q♣,K ♣,A♣,
cards =
2♥,3♥,4♥,5♥,6♥,7♥,8♥,9♥,10♥,J♥,Q♥,K ♥,A♥,
2♠,3♠,4♠,5♠,6♠,7♠,8♠,9♠,10♠,J♠,Q♠,K ♠,A♠
P1 {Q } = 4/52
P1 {♠} = 13/52
P1 {Q♠} = 1/52 = 4/52 × 13/52
|{z} | {z }
P1 {Q} P1 {♠}
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
Probability space 2 (unfair deck):
2♦,3♦,4♦,5♦,6♦,7♦,8♦,9♦,10♦,J♦,Q♦,K ♦,A♦,
2♥,3♥,4♥,5♥,6♥,7♥,8♥,9♥,10♥,J♥,Q♥,K ♥,A♥,
cards =
Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,
Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
Probability space 2 (unfair deck):
2♦,3♦,4♦,5♦,6♦,7♦,8♦,9♦,10♦,J♦,Q♦,K ♦,A♦,
2♥,3♥,4♥,5♥,6♥,7♥,8♥,9♥,10♥,J♥,Q♥,K ♥,A♥,
cards =
Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,
Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,
P2 {Q } = 28/52
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
Probability space 2 (unfair deck):
2♦,3♦,4♦,5♦,6♦,7♦,8♦,9♦,10♦,J♦,Q♦,K ♦,A♦,
2♥,3♥,4♥,5♥,6♥,7♥,8♥,9♥,10♥,J♥,Q♥,K ♥,A♥,
cards =
Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,
Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,
P2 {Q } = 28/52
P2 {♠} = 1/2
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
Probability space 2 (unfair deck):
2♦,3♦,4♦,5♦,6♦,7♦,8♦,9♦,10♦,J♦,Q♦,K ♦,A♦,
2♥,3♥,4♥,5♥,6♥,7♥,8♥,9♥,10♥,J♥,Q♥,K ♥,A♥,
cards =
Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,
Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,
P2 {Q } = 28/52
P2 {♠} = 1/2
P2 {Q♠} = 1/2
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
Probability space 2 (unfair deck):
2♦,3♦,4♦,5♦,6♦,7♦,8♦,9♦,10♦,J♦,Q♦,K ♦,A♦,
2♥,3♥,4♥,5♥,6♥,7♥,8♥,9♥,10♥,J♥,Q♥,K ♥,A♥,
cards =
Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,
Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,
P2 {Q } = 28/52
P2 {♠} = 1/2
P2 {Q♠} = 1/2 6= 28/52 × 1/2
| {z } |{z}
P2 {Q} P2 {♠}
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
An example in events independence
A card is drawn from a deck with 52 unknown cards.
Event 1: Q “queen”
Event 2: ♠ “spade”
Are these events independent?
Probability space 2 (unfair deck):
2♦,3♦,4♦,5♦,6♦,7♦,8♦,9♦,10♦,J♦,Q♦,K ♦,A♦,
2♥,3♥,4♥,5♥,6♥,7♥,8♥,9♥,10♥,J♥,Q♥,K ♥,A♥,
cards =
Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,
Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,Q♠,
P2 {Q } = 28/52
P2 {♠} = 1/2
P2 {Q♠} = 1/2 6= 28/52 × 1/2
| {z } |{z}
P2 {Q} P2 {♠}
Events are not independent.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 15 / 60
Further remarks on probability
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 16 / 60
Random variable
A collection of events in Ω is
mapped to an interval E on the
real line under such mapping.
Remark:
A random variable is a function from Ω to R, not a real number.
*Picture from http://theanalysisofdata.com/probability/2_1.htmll
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 17 / 60
Examples of random variables
G. Grimmett and D. Welsh, Probability: An Introduction. Oxford University Press, 2 edition, 2014.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 18 / 60
Probability distribution
FX (x) = P {X ≤ x} .
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 19 / 60
Probability density function
Remark:
Intuitively, pX (x) dx can be thought of as the probability of X
falling within the infinitesimal interval [x, x + dx].
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 20 / 60
Types of random variables
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 21 / 60
Function of random variable
The function of random variable Y = h (X ), for a random vari-
able X and measurable mapping h : R → R, is also a random
variable, which has its own probability distribution.
Example:
Let h(x) = x 2 and X : Ω → [1, 2] be a random variable such that
0 if x < 1
FX (x) = 1/2 if 1 ≤ x ≤ 2
1
if x > 2.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 23 / 60
Statistical moments
n o
The value E (X − b)k is called k-th moment around b of the ran-
dom variable X , for b ∈ R and k = 1, 2, · · · .
n k o
For b = E {X }, E X − E {X } is dubbed k-th central moment
of the random variable X .
n o
When b = 0, E X k is simply called k-th moment of the random
variable X .
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 24 / 60
Mean value
µX = Z {X }
E
= x dFX (x)
ZR
= x pX (x) dx.
R
Remark:
The mean value µX is the constant which best approximate the
random variable X . The error of this approximation is the standard
deviation σX .
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 25 / 60
Variance
The variance of the random variable X is defined as
n o
σX2 = E (X − µX )2
Z
= (x − µX )2 dFX (x)
ZR
= (x − µX )2 pX (x) dx.
R
Remark:
σX2 has the same unit as X 2 .
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 26 / 60
Standard deviation and variation coefficient
δX = σX /µX , µX 6= 0.
Remark:
σX has the same unit as X and δX is dimensionless.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 27 / 60
Sknewness
(measure of “tailedness”)
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 29 / 60
Entropy
which is equivalent to
Z
S (pX ) = − pX (x) ln pX (x) dx.
R
Remark: X
For discrete random variables S (PX ) = − Pk ln (Pk ).
k
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 30 / 60
Second-order random variables
In consequence,
E {X } < +∞,
and hence
n o 2
σX2 = E X 2 − E {X } < +∞.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 31 / 60
Probability Distributions
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 32 / 60
Binomial
Notation: B(n, p)
Support: {0, 1, 2, · · · , n}
Parameters:
n ∈ N — number of trials
p ∈ [0, 1] — success probability
PMF: Cumulative Distribution Function
n
Pk = p k (1 − p)n−k
k
Statistics:
µ = np
σ 2 = n p (1 − p)
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 33 / 60
Poisson
Probability Mass Function
Notation: Poisson(λ)
Support: {0, 1, 2, 3, · · · }
Parameter:
λ > 0 — mean
PMF: Cumulative Distribution Function
λk e −λ
Pk =
k!
Statistics:
µ=λ
σ2 = λ
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 34 / 60
Uniform
Notation: U(a, b)
Support: [a, b]
Parameters:
−∞ < a < b < +∞ — boundaries
PDF:
1 Cumulative Distribution Function
pX (x) = 1 (x)
b − a [a,b]
Statistics:
1
µ= (a + b)
2
1
σ2 = (b − a)2
12
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 35 / 60
Beta
Probability Density Function
Notation: Beta(α, β)
Support: [0, 1]
Parameters:
α — shape parameter
β — shape parameter
PDF:
Γ(α + β) α−1 Cumulative Distribution Function
pX (x) = x (1−x)β−1 1[0,1] (x)
Γ(α)Γ(β)
Statistics:
α
µ=
(α + β)
αβ
σ2 =
(α + β)2 (α + β + 1)
Beta distribution — Wikipedia, The Free Encyclopedia, 2017.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 36 / 60
Exponential
Probability Density Function
Notation: Exp(λ)
Support: [0, +∞)
Parameter:
λ — rate parameter
PDF:
Cumulative Distribution Function
Statistics:
µ = λ−1
σ 2 = λ−2
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 37 / 60
Gamma
Probability Density Function
Notation: Gamma(k, θ)
Support: (0, +∞)
Parameters:
k — shape parameter
θ — scale parameter
PDF:
Cumulative Distribution Function
1
pX (x) = x k−1 e (−x/θ) 1(0,+∞) (x)
Γ(k) θk
Statistics:
µ =kθ
σ2 = k θ2
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 38 / 60
Gaussian
Notation: N (µ, σ 2 )
Support: (−∞, +∞)
Parameters:
µ — mean
σ 2 — variance Cumulative Distribution Function
PDF:
( )
1 (x − µ)2
pX (x) = p exp −
2 π σ2 2 σ2
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 39 / 60
Log-normal
Notation: ln N (m, s 2 )
Support: (−∞, +∞)
Parameters:
m ∈ R — location parameter
s > 0 — scale parameter
PDF:
( ) Cumulative Distribution Function
1 (ln x − m)2
pX (x) = √ exp −
x 2 π s2 2 s2
Statistics:
2
µ = e m+s /2
2 2
σ 2 = (e s − 1) e 2 m+s
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 40 / 60
Cauchy
Probability Density Function
1 γ2
pX (x) =
πγ (x − x0 )2 + γ 2
Statistics:
µ = undefined
σ 2 = undefined (not 2nd-order RV)
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 41 / 60
Characterization of a probability distribution
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 42 / 60
Characterization of a probability distribution
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 42 / 60
Characterization of a probability distribution
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 42 / 60
Characterization of a probability distribution
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 42 / 60
Characterization of a probability distribution
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 42 / 60
Probability in Dimension n
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 43 / 60
Random vector
A collection of event in Ω
is mapped into a region on
the Euclidean space under such
mapping.
Thus,
Z b1 Z bn
P {a < X ≤ b} = ··· dFX (x1 , · · · , xn ),
a1 an
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 45 / 60
Joint probability density function
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 46 / 60
Marginal probability density function
Z +∞ Z +∞
pXi (xi ) = ··· pX (x1 , · · · , xn ) dx1 · · · dxi−1 dxi+1 · · · dxn ,
| −∞ {z −∞
}
n−1 times
for i = 1, · · · , n.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 47 / 60
Conditional distribution
FX ,Y (x, y )
FX |Y (x | y ) = .
FY (y )
Thus
FX ,Y (x, y ) = FX |Y (x|y ) × FY (y ),
and
pX ,Y (x, y ) = pX |Y (x|y ) × pY (y ).
Remark:
This definition extends naturally to the n-dimensional case.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 48 / 60
Independence of distributions
FX |Y (x|y ) = FX (x).
FX Y (x, y ) = FX (x) × FY (y ),
and
pX Y (x, y ) = pX (x) × pY (y ).
Remark:
This definition extends naturally to the n-dimensional case.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 49 / 60
Statistics of random vectors
second-order random vector
n o Z
E k X k2 = k x k2 dFX (x) < +∞
Rn
mean vector
Z
mX = E {X} = x dFX (x) ∈ Rn
Rn
correlation matrix
n o Z
[RXY ] = E XYT = xxT dFXY (x, y) ∈ Rn×n
Rn
covariance matrix
n o
[KXY ] = E (X − mX ) (Y − mY )T ∈ Rn×n
= [RXY ] − mX mT
Y
Remark:
Matrices [RXY ] and [KXY ] are symetric positive semi-definite
when X = Y.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 50 / 60
Correlation of random variables
[RXY ] = mX mT
Y.
independence =⇒ uncorrelation
But uncorrelated random vectors are not independent in general.
6 independence
uncorrelation =⇒
Remark:
Uncorrelated random vectors which the joint distribution is Gaussian
are independent.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 51 / 60
Notions of Random Processes
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 52 / 60
Random process
A real-valued random process (also called stochastic process) defined
on probability space (Ω, Σ, P), indexed by t ∈ T , is a mapping
(t, ω) ∈ T × Ω → X (t, ω) ∈ R,
such that, for fixed t, the output is a random variable X (t, ·), while
for fixed ω, X (·, ω) is a function of t (sample function).
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 54 / 60
Wiener process (Brownian motion process)
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 55 / 60
White noise and colored noise
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 56 / 60
Finite-dimensional distribution
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 57 / 60
Statistics of a random process
second-order random process
n o Z
E Xt2 = x 2 dFXt (x) < +∞
R
mean function
Z
µX (t) = E {Xt } = x dFXt (x)
R
correlation function
Z
corrX (t1 , t2 ) = E Xt1 Xt2 = x1 x2 dFXt1 Xt2 (x1 , x2 )
R
covariance function
n o
covX (t1 , t2 ) = E Xt1 − µX (t1 ) Xt2 − µX (t2 )
= corrX (t1 , t2 ) − µX (t1 )µX (t2 )
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 58 / 60
Stationary process
A stochastic process is stationary when all the random variables of
that stochastic process are identically distributed.
c A. Cunha Jr (UERJ) Modeling and Quantification of Uncertainties in Physical Systems 59 / 60
References
G. Grimmett and D. Welsh, Probability: An Introduction. Oxford University Press, 2 edition, 2014.
A. Papoulis and S. U. Pillai, Probability, Random Variables and Stochastic Processes. McGraw-Hill
Europe; 4th edition, 2002.