UNIT 5

BIDIMENSIONAL RANDOM VARIABLES

Bachelor’s Degree in Business Administration

University of Valladolid
Outline

5.1. Definition of bidimensional random variable

5.2. Discrete bidimensional random variables
5.3. Continuous bidimensional random variables
5.4. Characteristics of a bidimensional random variable
5.5. Sums of independent random variables

Unit 5. Bidimensional Random Variables 2 / 65

5.1. Definition of bidimensional random variable

• We study simultaneously two random variables to analyse the possible

relationship between them.

• Random experiment with sample space Ω

• X and Y two random variables defined in such space.

• (X , Y ) is a bidimensional random variable.

▷ (X , Y ) discrete bidimensional random variable if X and Y are
discrete.
▷ (X , Y ) continuous bidimensional random variable if X and Y are
continuous.

Unit 5. Bidimensional Random Variables 3 / 65

5.2. Discrete bidimensional random variables

Unit 5. Bidimensional Random Variables 4 / 65

5.2.1. Joint distribution
The joint probability function collects the probabilities of the values of a
discrete bidimensional variable.

p(X = xi , Y = yj ), for i = 1, 2, . . . , j = 1, 2, . . .

We distribute all the probability, 1, among a finite or countably infinite

number of values (points in the plane):

Unit 5. Bidimensional Random Variables 5 / 65

Properties of the joint probability function:

1. p(X = xi , Y = yj ) ≥ 0, for each value (xi , yj ) of the variable

∞ P
P ∞
2. p(X = xi , Y = yj ) = 1
i=1 j=1

Probability of an event:
P
p[(X , Y ) ∈ B] = p(X = xi , Y = yj )
(xi ,yj )∈B

Unit 5. Bidimensional Random Variables 6 / 65

Unit 5. Bidimensional Random Variables 7 / 65
Unit 5. Bidimensional Random Variables 8 / 65
Unit 5. Bidimensional Random Variables 9 / 65
Unit 5. Bidimensional Random Variables 10 / 65
5.2.2. Marginal distributions

From the joint probability function of (X , Y ), we can obtain the probability

functions of each of the two variables, which are called marginal probability
functions.

• Marginal probability function of X :

∞
P
p(X = xi ) = p(X = xi , Y = yj )
j=1

• Marginal probability function of Y :

∞
P
p(Y = yj ) = p(X = xi , Y = yj )
i=1

Unit 5. Bidimensional Random Variables 11 / 65

Unit 5. Bidimensional Random Variables 12 / 65
Unit 5. Bidimensional Random Variables 13 / 65
Unit 5. Bidimensional Random Variables 14 / 65
Unit 5. Bidimensional Random Variables 15 / 65
Joint and marginal distributions:

From the marginal distributions, we can obtain marginal expectations and

variances of each of the variables.

Unit 5. Bidimensional Random Variables 16 / 65

From the marginal distributions, we can obtain marginal expectations and
variances of each of the variables.

• Marginal expected values:

∞
P ∞
P
E (X ) = xi · p(X = xi ) E (Y ) = yj · p(Y = yj )
i=1 j=1

• Marginal variances:

Var (X ) = E (X 2 ) − [E (X )]2 Var (Y ) = E (Y 2 ) − [E (Y )]2

Unit 5. Bidimensional Random Variables 17 / 65

Unit 5. Bidimensional Random Variables 18 / 65
E (X ) = 1 · (3/10) + 2 · (6/10) + 3 · (1/10) = 1.8
E (X 2 ) = 12 · (3/10) + 22 · (6/10) + 32 · (1/10) = 3.6
Var (X ) = E (X 2 ) − [E (X )]2 = 3.6 − 1.82 = 0.36

Unit 5. Bidimensional Random Variables 18 / 65

Unit 5. Bidimensional Random Variables 19 / 65
E (Y ) = 0 · (6/10) + 1 · (3/10) + 2 · (1/10) = 0.5
E (Y 2 ) = 02 · (6/10) + 12 · (3/10) + 22 · (1/10) = 0.7
Var (Y ) = E (Y 2 ) − [E (Y )]2 = 0.7 − 0.52 = 0.45

Unit 5. Bidimensional Random Variables 19 / 65

5.2.3. Conditional distributions

▷ Conditional probability function of X given (Y = yj ):

p(X =x,Y =yj )
p(X = x|Y = yj ) = p(Y =yj ) , for all x ∈ R

▷ Conditional probability function of Y given (X = xi ):

p(X =xi ,Y =y )
p(Y = y |X = xi ) = p(X =xi ) , for all y ∈ R

Unit 5. Bidimensional Random Variables 20 / 65

Example 1: Conditional distribution of Y given X = 1

Unit 5. Bidimensional Random Variables 21 / 65

Example 1: Conditional distribution of Y given X = 1

p(X =1,Y =0) 1/10

p(Y = 0|X = 1) = p(X =1) = 3/10 = 1/3
p(X =1,Y =1) 1/10
p(Y = 1|X = 1) = p(X =1) = 3/10 = 1/3
p(X =1,Y =2) 1/10
p(Y = 2|X = 1) = p(X =1) = 3/10 = 1/3

THE PROBABILITIES SUM 1

Unit 5. Bidimensional Random Variables 21 / 65

In the same way we obtain the conditional distribution of Y given X = 2:

Unit 5. Bidimensional Random Variables 22 / 65

And the conditional distribution of Y given X = 3:

Unit 5. Bidimensional Random Variables 23 / 65

Sometimes we only know that a certain event has occurred, which we can
call event B. In this case, the calculations are performed in the same way:

Unit 5. Bidimensional Random Variables 24 / 65

For instance, we can calculate the conditional probability that Y = 1 given
X ≥ 2:

Unit 5. Bidimensional Random Variables 25 / 65

For instance, we can calculate the conditional probability that Y = 1 given
X ≥ 2:

Unit 5. Bidimensional Random Variables 25 / 65

From conditional distributions, one can obtain the conditional expecta-
tions and conditional variances:

▷ Conditional expectation and variance of X given (Y = yj ):

∞
P
E (X |Y = yj ) = xi · p(X = xi |Y = yj )
i=1

Var (X |Y = yj ) = E (X 2 |Y = yj ) − [E (X |Y = yj )]2

▷ Conditional expectation and variance of Y given (X = xi ):

∞
P
E (Y |X = xi ) = yj · p(Y = yj |X = xi )
j=1

Var (Y |X = xi ) = E (Y 2 |X = xi ) − [E (Y |X = xi )]2

Unit 5. Bidimensional Random Variables 26 / 65

Unit 5. Bidimensional Random Variables 27 / 65
E (Y |X = 1) = 0 · p(Y = 0|X = 1) + 1 · p(Y = 1|X = 1)+
+ 2 · p(Y = 2|X = 1) = 0 · 1/3 + 1 · 1/3 + 2 · 1/3 = 1

E (Y |X = 2) = 0 · p(Y = 0|X = 2) + 1 · p(Y = 1|X = 2)+

+ 2 · p(Y = 2|X = 2) = 0 · 2/3 + 1 · 1/3 + 2 · 0 = 1/3

E (Y |X = 3) = 0 · p(Y = 0|X = 3) + 1 · p(Y = 1|X = 3)+

+ 2 · p(Y = 2|X = 3) = 0 · 1 + 1 · 0 + 2 · 0 = 0

Unit 5. Bidimensional Random Variables 27 / 65

E (Y |X = x) is a function of x:

Doing the same for X conditioning on the different values of Y we obtain:

Similarly, we can obtain the conditional variances Var (X |Y = y ) and

Var (Y |X = x).

Unit 5. Bidimensional Random Variables 28 / 65

Property of the conditional expectation: law of iterated expectations

E (Y |X = x) is a function of x.
Then, g (X ) = E (Y |X ) is a transformation of X and, therefore, is another
unidimensional random variable.
It can be shown that:

E (E (Y |X )) = E (Y )

Unit 5. Bidimensional Random Variables 29 / 65

Property of the conditional expectation: law of iterated expectations

Unit 5. Bidimensional Random Variables 30 / 65

Important remark

The joint probability function can be written as:

p(X = xi , Y = yj ) = p(X = xi ) · p(Y = yj |X = xi ) =

= p(Y = yj ) · p(X = xi |Y = yj )

Therefore, the probability function can be calculated by knowing the marginal

of one variable and the conditional of the other by the first one.

Unit 5. Bidimensional Random Variables 31 / 65

5.2.4. Independence of random variables

Two equivalent definitions of independence

Unit 5. Bidimensional Random Variables 32 / 65

5.2.4. Independence of random variables

Two equivalent definitions of independence

First definition: X and Y are independent if

p(X = xi , Y = yj ) = p(X = xi ) · p(Y = yj ), i = 1, 2, . . . , j = 1, 2, . . .

Unit 5. Bidimensional Random Variables 32 / 65

5.2.4. Independence of random variables

Two equivalent definitions of independence

First definition: X and Y are independent if

p(X = xi , Y = yj ) = p(X = xi ) · p(Y = yj ), i = 1, 2, . . . , j = 1, 2, . . .

Second definition: X and Y are independent if

p(X =xi ,Y =yj )

p(X = xi |Y = yj ) = p(Y =yj ) = p(X = xi )

p(X =xi ,Y =yj )

p(Y = yj |X = xi ) = p(X =xi ) = p(Y = yj )

Unit 5. Bidimensional Random Variables 32 / 65

Example 1.
Joint and marginal distributions:

18 1
p(X = 1) · p(Y = 0) = (3/10) · (6/10) = 100 ̸= p(X = 1, Y = 0) = 10

Hence, X and Y are not independent.

Unit 5. Bidimensional Random Variables 33 / 65

Moreover, if (X , Y ) is a discrete bidimensional random variable with X and
Y independent, then the random variables U = g (X ) and V = h(Y ) are
also independent.

Unit 5. Bidimensional Random Variables 34 / 65

5.3. Continuous bidimensional random variables

(X , Y ) is a continuous bidimensional random variable if the random variables

X and Y are continuous.

Example 3:
X = monthly disposable income
Y = consumption of perishable goods

Unit 5. Bidimensional Random Variables 35 / 65

5.3.1. Joint distribution
The joint density function: f(X ,Y ) (x, y )

We distribute the entire probability continuously over a subset of R2 .

Unit 5. Bidimensional Random Variables 36 / 65

Example 3:

Unit 5. Bidimensional Random Variables 37 / 65

5.3.2. Marginal distributions

If a bidimensional random variable, (X , Y ), is continuous, X and Y are

continuous, and their marginal density functions can be obtained, similarly
to the discrete case, but by integrating the joint density function:

• Marginal density function of X :

• Marginal density function of Y :

Unit 5. Bidimensional Random Variables 38 / 65

Example 3:

Unit 5. Bidimensional Random Variables 39 / 65

Example 3:

Unit 5. Bidimensional Random Variables 39 / 65

Ejemplo 3:

Unit 5. Bidimensional Random Variables 40 / 65

Ejemplo 3:

Unit 5. Bidimensional Random Variables 40 / 65

From the marginal distributions, the marginal expected values and vari-
ances of each variable can be obtained.

• Marginal expected value and variance of X :

• Marginal expected value and variance of Y :

Unit 5. Bidimensional Random Variables 41 / 65

Example 3:

From the density function of X , we can obtain its expected value and vari-
ance:

Unit 5. Bidimensional Random Variables 42 / 65

Example 3:

From the density function of X , we can obtain its expected value and vari-
ance:

Unit 5. Bidimensional Random Variables 42 / 65

Example 3:

From the density function of Y , we can obtain its expected value and vari-
ance:

Unit 5. Bidimensional Random Variables 43 / 65

Example 3:

From the density function of Y , we can obtain its expected value and vari-
ance:

Unit 5. Bidimensional Random Variables 43 / 65

5.3.3. Conditional distributions
Let (X , Y ) be a continuous random variable and let y be a value such that
fY (y ) ̸= 0. The conditional density function of X given the event
(Y = y) is the function

In the same way, given a value, x, such that fX (x) ̸= 0, the conditional
density function of Y given the event (X = x) is the function

Unit 5. Bidimensional Random Variables 44 / 65

Example 3:

Unit 5. Bidimensional Random Variables 45 / 65

Example 3:

Only defined for x > 0 (for fX (x) > 0), in which case, is defined as:

Unit 5. Bidimensional Random Variables 45 / 65

Particular cases:
If x = 4:

Unit 5. Bidimensional Random Variables 46 / 65

Particular cases:
If x = 4:

Conditional probabilities calculation:

Unit 5. Bidimensional Random Variables 46 / 65

Using the corresponding conditional density functions, we can calculate con-
ditional expected values and variances:
Specifically, the moments of Y conditioned on the event (X = x) are cal-
culated as:

Unit 5. Bidimensional Random Variables 47 / 65

Example 3:
Using the respective conditional density functions, one can obtain the aver-
age consumption and variance for different levels of disposable income:

Unit 5. Bidimensional Random Variables 48 / 65

What is the expected consumption of perishable goods for a family with an
income of 2500 euros?
For X = 2.5:

Unit 5. Bidimensional Random Variables 49 / 65

Property of the conditional expected value: law of iterated expecta-
tions
E (Y |X = x) is a function of x. Hence, g (X ) = E (Y |X ) is a transformation
of the random variable X and thus another one-dimensional random variable.
We can then calculate its expected value, and it can be shown that

E (E (Y |X )) = E (Y )

Unit 5. Bidimensional Random Variables 50 / 65

Property of the conditional expected value: law of iterated expecta-
tions
E (Y |X = x) is a function of x. Hence, g (X ) = E (Y |X ) is a transformation
of the random variable X and thus another one-dimensional random variable.
We can then calculate its expected value, and it can be shown that

E (E (Y |X )) = E (Y )

Unit 5. Bidimensional Random Variables 50 / 65

Important note:

The joint density function can be written as:

Hence, we can calculate the joint density function knowing the marginal of
a variable and the conditional of the other variable given this one.

Unit 5. Bidimensional Random Variables 51 / 65

5.3.4. Independence of random variables
Two equivalent definitions of independence

Unit 5. Bidimensional Random Variables 52 / 65

5.3.4. Independence of random variables
Two equivalent definitions of independence

First definition: X and Y are independent if

Unit 5. Bidimensional Random Variables 52 / 65

5.3.4. Independence of random variables
Two equivalent definitions of independence

First definition: X and Y are independent if

Second definition: X and Y are independent if

Unit 5. Bidimensional Random Variables 52 / 65

Unit 5. Bidimensional Random Variables 53 / 65
5.4. Characteristics of a bidimensional random
variable
From the joint distribution, one can obtain expected values of functions
of a bidimensional variable:

Unit 5. Bidimensional Random Variables 54 / 65

5.4. Characteristics of a bidimensional random
variable
From the joint distribution, one can obtain expected values of functions
of a bidimensional variable:

Properties of the expected value:

1.- Linearity: E (a · X + b · Y + c) = a · E (X ) + b · E (Y ) + c
In general, E (a1 · X1 + · · · + an · Xn + c) = a1 · E (X1 ) + . . . an · E (Xn ) + c

Unit 5. Bidimensional Random Variables 54 / 65

5.4. Characteristics of a bidimensional random
variable
From the joint distribution, one can obtain expected values of functions
of a bidimensional variable:

Properties of the expected value:

1.- Linearity: E (a · X + b · Y + c) = a · E (X ) + b · E (Y ) + c
In general, E (a1 · X1 + · · · + an · Xn + c) = a1 · E (X1 ) + . . . an · E (Xn ) + c
2.- E (X + Y ) = E (X ) + E (Y )
In general, E (X1 + · · · + Xn ) = E (X1 ) + · · · + E (Xn )

Unit 5. Bidimensional Random Variables 54 / 65

5.4. Characteristics of a bidimensional random
variable
From the joint distribution, one can obtain expected values of functions
of a bidimensional variable:

Properties of the expected value:

1.- Linearity: E (a · X + b · Y + c) = a · E (X ) + b · E (Y ) + c
In general, E (a1 · X1 + · · · + an · Xn + c) = a1 · E (X1 ) + . . . an · E (Xn ) + c
2.- E (X + Y ) = E (X ) + E (Y )
In general, E (X1 + · · · + Xn ) = E (X1 ) + · · · + E (Xn )
3.- If X and Y are independent ⇒ E (X · Y ) = E (X ) · E (Y )
The reciprocal result is not true!
Unit 5. Bidimensional Random Variables 54 / 65
Covariance:

σXY = Cov (X , Y ) = E [(X − E (X ))(Y − E (Y ))]

Interpretation: measure of linear association between X e Y

• σXY > 0 ⇒ large values of X ↔ large values of Y

• σXY < 0 ⇒ large values of X ↔ small values of Y

Unit 5. Bidimensional Random Variables 55 / 65

Covariance:

σXY = Cov (X , Y ) = E [(X − E (X ))(Y − E (Y ))]

Interpretation: measure of linear association between X e Y

• σXY > 0 ⇒ large values of X ↔ large values of Y

• σXY < 0 ⇒ large values of X ↔ small values of Y

Properties of the covariance:

P1. Cov (X , Y ) = E (X · Y ) − E (X ) · E (Y )

Unit 5. Bidimensional Random Variables 55 / 65

Covariance:

σXY = Cov (X , Y ) = E [(X − E (X ))(Y − E (Y ))]

Interpretation: measure of linear association between X e Y

• σXY > 0 ⇒ large values of X ↔ large values of Y

• σXY < 0 ⇒ large values of X ↔ small values of Y

Properties of the covariance:

P1. Cov (X , Y ) = E (X · Y ) − E (X ) · E (Y )
P2. Cov (X , X ) = Var (X )

Unit 5. Bidimensional Random Variables 55 / 65

Covariance:

σXY = Cov (X , Y ) = E [(X − E (X ))(Y − E (Y ))]

Interpretation: measure of linear association between X e Y

• σXY > 0 ⇒ large values of X ↔ large values of Y

• σXY < 0 ⇒ large values of X ↔ small values of Y

Properties of the covariance:

P1. Cov (X , Y ) = E (X · Y ) − E (X ) · E (Y )
P2. Cov (X , X ) = Var (X )
P3. Cov (X , Y ) = Cov (Y , X )

Unit 5. Bidimensional Random Variables 55 / 65

Covariance:

σXY = Cov (X , Y ) = E [(X − E (X ))(Y − E (Y ))]

Interpretation: measure of linear association between X e Y

• σXY > 0 ⇒ large values of X ↔ large values of Y

• σXY < 0 ⇒ large values of X ↔ small values of Y

Properties of the covariance:

P1. Cov (X , Y ) = E (X · Y ) − E (X ) · E (Y )
P2. Cov (X , X ) = Var (X )
P3. Cov (X , Y ) = Cov (Y , X )
P4. If X and Y are independent ⇒ Cov (X , Y ) = 0

Unit 5. Bidimensional Random Variables 55 / 65

Covariance:

σXY = Cov (X , Y ) = E [(X − E (X ))(Y − E (Y ))]

Interpretation: measure of linear association between X e Y

• σXY > 0 ⇒ large values of X ↔ large values of Y

• σXY < 0 ⇒ large values of X ↔ small values of Y

Properties of the covariance:

P1. Cov (X , Y ) = E (X · Y ) − E (X ) · E (Y )
P2. Cov (X , X ) = Var (X )
P3. Cov (X , Y ) = Cov (Y , X )
P4. If X and Y are independent ⇒ Cov (X , Y ) = 0
P5. Cov (a + b · X , c + d · Y ) = bd · Cov (X , Y ) ⇒ The covariance varies
under changes of scale.

Unit 5. Bidimensional Random Variables 55 / 65

Linear correlation coefficient:

Cov (X ,Y )
ρXY = σX ·σY

• sign(ρXY ) = sign(σXY )
• −1 ≤ ρXY ≤ 1

Unit 5. Bidimensional Random Variables 56 / 65

Linear correlation coefficient:

Cov (X ,Y )
ρXY = σX ·σY

• sign(ρXY ) = sign(σXY )
• −1 ≤ ρXY ≤ 1

Properties of the linear correlation coefficient:

P1. If X and Y are independent ⇒ ρXY = 0 (the contrary is not necessarily

true)

Unit 5. Bidimensional Random Variables 56 / 65

Linear correlation coefficient:

Cov (X ,Y )
ρXY = σX ·σY

• sign(ρXY ) = sign(σXY )
• −1 ≤ ρXY ≤ 1

Properties of the linear correlation coefficient:

P1. If X and Y are independent ⇒ ρXY = 0 (the contrary is not necessarily

true)

P2. ρ(a + b · X , c + d · Y ) = ρ(X , Y ) whenever sign(b) = sign(d)

Unit 5. Bidimensional Random Variables 56 / 65

Linear correlation coefficient:

Cov (X ,Y )
ρXY = σX ·σY

• sign(ρXY ) = sign(σXY )
• −1 ≤ ρXY ≤ 1

Properties of the linear correlation coefficient:

P1. If X and Y are independent ⇒ ρXY = 0 (the contrary is not necessarily

true)

P2. ρ(a + b · X , c + d · Y ) = ρ(X , Y ) whenever sign(b) = sign(d)

Hence, ρXY is invariant under changes of scale

Unit 5. Bidimensional Random Variables 56 / 65

Linear correlation coefficient:

Cov (X ,Y )
ρXY = σX ·σY

• sign(ρXY ) = sign(σXY )
• −1 ≤ ρXY ≤ 1

Properties of the linear correlation coefficient:

P1. If X and Y are independent ⇒ ρXY = 0 (the contrary is not necessarily

true)

P2. ρ(a + b · X , c + d · Y ) = ρ(X , Y ) whenever sign(b) = sign(d)

Hence, ρXY is invariant under changes of scale

ρXY measures sign and intensity of the linear relationship between X and
Y

Unit 5. Bidimensional Random Variables 56 / 65

Example 1 (discrete variables):

We know: E (X ) = 1.8; Var (X ) = 0.36; E (Y ) = 0.5; Var (Y ) = 0.45

Unit 5. Bidimensional Random Variables 57 / 65

Example 1 (discrete variables):

We know: E (X ) = 1.8; Var (X ) = 0.36; E (Y ) = 0.5; Var (Y ) = 0.45

To calculate the covariance, we calculate previously E (X · Y ):

Unit 5. Bidimensional Random Variables 57 / 65

Example 1 (discrete variables):
Then, we calculate the covariance:

Unit 5. Bidimensional Random Variables 58 / 65

Example 1 (discrete variables):
Then, we calculate the covariance:

We can calculate the linear correlation coefficient:

This value indicates a weak negative linear relationship between the vari-
ables

Unit 5. Bidimensional Random Variables 58 / 65

Example 3 (continuous variables):

We know: E (X ) = 2; Var (X ) = 2; E (Y ) = 1; Var (Y ) = 1

Unit 5. Bidimensional Random Variables 59 / 65

Example 3 (continuous variables):

We know: E (X ) = 2; Var (X ) = 2; E (Y ) = 1; Var (Y ) = 1

To calculate the covariance, we calculate previously E (X · Y ):

Unit 5. Bidimensional Random Variables 59 / 65

Example 3 (continuous variables):
Then, we calculate the covariance:

Unit 5. Bidimensional Random Variables 60 / 65

Example 3 (continuous variables):
Then, we calculate the covariance:

We can calculate the linear correlation coefficient:

This value indicates a certain positive linear relationship between the

variables.

Unit 5. Bidimensional Random Variables 60 / 65

Variance of “sums” of variables:

P6. Var (a · X + b · Y + c) = a2 · Var (X ) + b 2 · Var (Y ) + 2 · ab · Cov (X , Y )

Particular case:
Var (X + Y ) = Var (X ) + Var (Y ) + 2 · Cov (X , Y )
Particular case:
Var (X − Y ) = Var (X ) + Var (Y ) − 2 · Cov (X , Y )

Unit 5. Bidimensional Random Variables 61 / 65

 Variance of “sums” of variables:

P6. Var (a · X + b · Y + c) = a2 · Var (X ) + b 2 · Var (Y ) + 2 · ab · Cov (X , Y )

Particular case:
Var (X + Y ) = Var (X ) + Var (Y ) + 2 · Cov (X , Y )
Particular case:
Var (X − Y ) = Var (X ) + Var (Y ) − 2 · Cov (X , Y )

P67. If X and Y are independent:

⇒ Var (X + Y ) = Var (X ) + Var (Y )
⇒ Var (X − Y ) = Var (X ) + Var (Y )
In general, if X1 , . . . , Xn are independent:
Var (X1 + · · · + Xn ) = Var (X1 ) + · · · + Var (Xn )
Var (a1 · X1 + · · · + an Xn ) = (a1 )2 · Var (X1 ) + · · · + (an )2 · Var (Xn )

Unit 5. Bidimensional Random Variables 61 / 65

5.5. Sums of independent random variables
• Bernoulli
X1 , . . . , Xn n Bernoulli independent random variables, with the same
probability of success, p

X1 + · · · + Xn → B(n, p)

Unit 5. Bidimensional Random Variables 62 / 65

5.5. Sums of independent random variables
• Bernoulli
X1 , . . . , Xn n Bernoulli independent random variables, with the same
probability of success, p

X1 + · · · + Xn → B(n, p)

• Binomial
X1 , . . . , Xk independent random variables with distribution B(ni , p),
i = 1, . . . , k

X1 + · · · + Xk → B(n1 + · · · + nk , p)

Unit 5. Bidimensional Random Variables 62 / 65

5.5. Sums of independent random variables
• Bernoulli
X1 , . . . , Xn n Bernoulli independent random variables, with the same
probability of success, p

X1 + · · · + Xn → B(n, p)

• Binomial
X1 , . . . , Xk independent random variables with distribution B(ni , p),
i = 1, . . . , k

X1 + · · · + Xk → B(n1 + · · · + nk , p)

• Poisson
X1 , . . . , Xn independent random variables with distribution P(λi ), i =
1, . . . , n

X1 + · · · + Xn → P(λ1 + · · · + λn )
Unit 5. Bidimensional Random Variables 62 / 65
• Geometric
X1 , . . . , Xr independent random variables with Geometric distribution
G (p)

X1 + · · · + Xr → BN(r , p)

Unit 5. Bidimensional Random Variables 63 / 65

• Geometric
X1 , . . . , Xr independent random variables with Geometric distribution
G (p)

X1 + · · · + Xr → BN(r , p)

• Negative Binomial
X1 , . . . , Xn independent random variables with distribution BN(ri , p),
i = 1, . . . , n

X1 + · · · + Xn → BN(r1 + · · · + rn , p)

Unit 5. Bidimensional Random Variables 63 / 65

• Normal
X1 , . . . , Xn independent random variables with distribution N(µi , σi ),
i = 1, . . . , n
q
X1 + · · · + Xn → N(µ1 + · · · + µn , σ12 + · · · + σn2 )
q
a1 ·X1 +· · ·+an ·Xn → N(a1 ·µ1 +· · ·+an ·µn , a12 · σ12 + · · · + an2 · σn2 )

Unit 5. Bidimensional Random Variables 64 / 65

References

▷ NEWBOLD, P., CARLSON, W., and THORNE, B. (2022): Statistics

for Business and Economics, 10th Global Edition. Ed. Pearson. Chap-
ters 4 and 5

▷ ROSS, S. (2023): A First Course in Probability, 10th Global Edition.

Ed. Pearson. Chapters 6 and 7

Unit 5. Bidimensional Random Variables 65 / 65