8mmpjoint Annotated

Multivariate Distributions
Giulia Giantesio
giulia.giantesio@unicatt.it
Introduction
Definition (Probability function)

Given a σ-algebra family of events A on a sample space Ω, a function p : A → [0, 1] is called a probability
measure on Ω if
1 p(A) ⩾ 0 ∀A ∈ A;
2 p(Ω) = 1;
3 For any countable set of events, A1 , . . . , An , · · · ∈ A which are disjoint, we have
∞
! ∞
[ X
p Ai = p(Ai ).
i=1 i=1
a) p (∅) = 0;
b) p (Ac ) = 1 − p (A) ;
c) if A, B ∈ A, then p (A ∪ B) = p (A) + p (B) − p (A ∩ B) ;
d) if A ⊆ B, then p (A) ⩽ p (B).
Introduction
Definition (Conditional probability)

Let (Ω, A, p) be a probability space and take two events A, B ∈ A. We call
 p (A ∩ B) if p (B) > 0

p (A | B) = p (B)
p(A) otherwise

the conditional probability of A given B.
Definition
Two events A and B are called independent if and only if p(A ∩ B) = p(A)p(B).
Introduction
Definition (Random variable)

Let (Ω, A) a measurable space. A function X : Ω → R such that ∀x ∈ R : {ω ∈ Ω : X (ω) ≤ x} ∈ A is called
random variable.
Definition (Cumulative distribution function)

The cumulative distribution function (c.d.f.) FX : R → [0, 1] of the random variable X is defined by
FX (x) = p ({ω ∈ Ω : X (ω) ⩽ x}) = p ([X ⩽ x]) .
Introduction
Definition (Absolutely continuous rv)

Definition (Discrete rv)
For a nondiscrete absolutely continuous random variable
For a discrete random variable X , the X , its distribution function can be represented as
function fX : R → [0, 1] given by Z x
FX (x) = fX (t) dt

fX (x) = p [X (ω) = x] = p X −1 (x) −∞
is called probability mass function. where fX : R → R is called the probability density

function.
Multivariate Distributions
Giulia Giantesio
giulia.giantesio@unicatt.it
Joint Probability Mass Function
Definition (Joint Probability Mass Function)

If X and Y are two discrete random variables, we define the joint probability mass function by
fXY (x, y ) = p(X = x ∩ Y = y ),
where
▶ fXY (x, y ) ⩾ 0;
XX
▶ fXY (x, y ) = 1.
x y
Joint Probability Mass Function
Notation:
f (x, y ) = fXY (x, y ) = p(X = x ∩ Y = y ) = p(X = x, Y = y )
We can define the joint range for X and Y as
RXY = {(x, y )|f (x, y ) > 0}.
So that RXY ⊆ RX × RY = {(xi , yj )|xi ∈ RX , yj ∈ RY }.

Marginal PMFs
The joint PMF contains all the information regarding the distributions of X and Y .
Definition
We define marginal probability mass functions of X and Y :
X
fX (x) = f (x, yj ), for any x ∈ RX ,
yj ∈RY
X
fY (y ) = f (xi , y ), for any y ∈ RY .
xi ∈RX
Joint probability table
It represents the joint PMF for X and Y :
y1 y2 ... yn
x1 f (x1 , y1 ) f (x1 , y2 ) ... f (x1 , yn ) fX (x1 )
x2 f (x2 , y1 ) f (x2 , y2 ) ... f (x2 , yn ) fX (x2 )
.. .. .. .. ..
. . . . .
xm f (xm , y1 ) f (xm , y2 ) ... f (xm , yn ) fX (xm )
fY (y1 ) fY (y2 ) ... fY (yn ) 1
Remark
The probability that X = xi (Y = yk ) is obtained by adding all entries in the row (column)
corresponding to xi (yk ).
Joint cumulative distribution function
Definition
The joint cumulative distribution function of two discrete rv X and Y is defined by
XX
F (x, y ) = p(X ⩽ x, Y ⩽ y ) = f (u, v ).
u⩽x v ⩽y
Joint cumulative distribution function: properties
We define the marginal CDFs of X and Y :
FX (x) = lim F (x, y ), for any x,

y →∞
FY (y ) = lim F (x, y ), for any y .

x→∞
We must have
F (∞, ∞) = 1,
F (−∞, y ) = 0, for any y ,
F (x, −∞) = 0, for any x.
Joint cumulative distribution function: properties
Lemma
For two random variables X and Y , and real numbers x1 ⩽ x2 , y1 ⩽ y2 , we have
p(x1 < X ⩽ x2 , y1 < X ⩽ y2 ) = F (x2 , y2 ) − F (x1 , y2 ) − F (x2 , y1 ) + F (x1 , y1 ).

Example
Consider two random variables X and Y with joint PMF given in
Y =0 Y =1 Y =2
1 1 1
X =0 6 4 8
1 1 1
X =1 8 6 6
▶ Find p(X = 0, Y ⩽ 1);

Example
Consider two random variables X and Y with joint PMF given in
Y =0 Y =1 Y =2
1 1 1
X =0 6 4 8
1 1 1
X =1 8 6 6
▶ Find the marginal PMFs of X and Y .

Multinomial distribution
The multinomial distribution arises from an extension of the binomial experiment to situations
where each trial has k ⩾ 2 possible outcomes.
Suppose that we have an experiment with n independent trials, where each trial produces
exactly one of the events A1 , A2 , . . . , Ak (these events are mutually exclusive).
Moreover, suppose that each trial Ai occurs with probability pi .
We must have p1 + p2 + · · · + pk = 1.
Let Xi be the number of trials in which Ai occurs, then Xi is a binomial rv.
n!
p(X1 = n1 , X2 = n2 , . . . , Xk = nk ) = p1n1 p2n2 . . . pknk .
n1 !n2 ! . . . nk !
Multinomial distribution:example
Suppose that a fair dice is rolled 9 times. Compute the probability that 1 appears three times,
2 and 3 twice each, 4 and 5 once each, and 6 not at all.
Continuous case
Definition (Joint Probability Function)

If X and Y are two continuous random variables, we define the joint probability function or
joint density function by
Z Z
fXY (x, y )dxdy = p[(X , Y ) ∈ A], A ⊆ R2 ,
A
where
▶ fXY (x, y ) ⩾ 0;
Z +∞ Z +∞
▶ fXY (x, y )dxdy = 1.
−∞ −∞
Continuous case
Z b Z d
p(a < X < b, c < Y < d) = f (x, y )dxdy
a c
Continuous case: example
Let X and Y be two jointly continuous random variables with joint PDF
(
x + cy 2 if 0 ⩽ x ⩽ 1, 0 ⩽ y ⩽ 1
f (x, y ) =
0 otherwise
find the constant c.

Joint cumulative distribution Function
We define the joint cumulative distribution function:

Z x Z y
F (x, y ) = p(X ⩽ x, Y ⩽ y ) = f (u, v )dudv
−∞ −∞
∂2F
= f (x, y ).
∂x∂y
Marginal functions
Marginal distributions:
Z x Z ∞
FX (x) = f (u, v )dudv , for any x ∈ RX ,
−∞ −∞
Z ∞Z y
FY (y ) = f (u, v )dudv , for any y ∈ RY .
−∞ −∞
Marginal densities:
Z ∞
fX (x) = f (x, v )dv , for any x ∈ RX ,
−∞
Z ∞
fY (y ) = f (u, y )du, for any y ∈ RY .
−∞
Extension to the case of n variables: JCDF
FX1 X2 ...Xn (x1 , x2 , . . . , xn ) = F− →

−
→ ( x ) = p(X1 ⩽ x1 , X2 ⩽ x2 , . . . , Xn ⩽ xn ).
X
Extension to the case of n variables: JPMF discrete case
n
!
→
−
\
f−
→( x ) = p
X
[Xk = xk ]
k=1
fX1 X2 ...Xn (x1 , x2 , . . . , xn ) = p(X1 = x1 , X2 = x2 , . . . , Xn = xn )

Extension to the case of n variables: JPDF continuous case
Z Z Z
p[(X1 , X2 , . . . , Xn ) ∈ A] = ··· f (x1 , x2 , . . . , xn )dx1 . . . dxn .
A

8mmpjoint Annotated

Uploaded by

Copyright:

Available Formats

8mmpjoint Annotated

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

8mmpjoint Annotated

Uploaded by

Copyright:

Available Formats

Multivariate Distributions

Definition (Probability function)

Definition (Conditional probability)

the conditional probability of A given B.

Definition (Random variable)

Definition (Cumulative distribution function)

Definition (Absolutely continuous rv)

is called probability mass function. where fX : R → R is called the probability density

Definition (Joint Probability Mass Function)

fXY (x, y ) = p(X = x ∩ Y = y ),

f (x, y ) = fXY (x, y ) = p(X = x ∩ Y = y ) = p(X = x, Y = y )

We can define the joint range for X and Y as

RXY = {(x, y )|f (x, y ) > 0}.

So that RXY ⊆ RX × RY = {(xi , yj )|xi ∈ RX , yj ∈ RY }.

It represents the joint PMF for X and Y :

We define the marginal CDFs of X and Y :

FX (x) = lim F (x, y ), for any x,

FY (y ) = lim F (x, y ), for any y .

p(x1 < X ⩽ x2 , y1 < X ⩽ y2 ) = F (x2 , y2 ) − F (x1 , y2 ) − F (x2 , y1 ) + F (x1 , y1 ).

Consider two random variables X and Y with joint PMF given in

▶ Find p(X = 0, Y ⩽ 1);

Consider two random variables X and Y with joint PMF given in

▶ Find the marginal PMFs of X and Y .

Definition (Joint Probability Function)

find the constant c.

We define the joint cumulative distribution function:

FX1 X2 ...Xn (x1 , x2 , . . . , xn ) = F− →

fX1 X2 ...Xn (x1 , x2 , . . . , xn ) = p(X1 = x1 , X2 = x2 , . . . , Xn = xn )

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.