JOINT DISTRIBUTION & TYPES OF

DISTRIBUTIONS

GHULAM MUSTAFA

February 20, 2026

Definition of Joint Probability Distribution

Definition
A joint probability distribution describes the probability of two or
more random variables occurring simultaneously. It provides a
complete description of the probabilistic relationship between
multiple random variables defined on the same sample space.
Joint Probability Mass Function (PMF)
For two discrete random variables X and Y :
Joint PMF

p(x, y ) = P(X = x and Y = y )

Properties:
▶ p(x, y ) ≥ 0 for all (x, y )
XX
▶ p(x, y ) = 1
x y

Example
For rolling two fair dice:
1
p(x, y ) = for all x, y ∈ {1, 2, 3, 4, 5, 6}
36
Example: Rolling Two Fair Dice
Solution:
▶ Let X = outcome of first die, Y = outcome of second die.
The sample space has 36 equally likely outcomes.
▶ Joint Probability Mass Function (PMF):
1
pX ,Y (x, y ) = P(X = x, Y = y ) = , x, y = 1, . . . , 6.
36
▶ Marginal PMFs:
6 6
X X 1 6 1
pX (x) = pX ,Y (x, y ) = = = , x = 1, . . . , 6.
36 36 6
y =1 y =1
1
Similarly, pY (y ) = for y = 1, . . . , 6. So each die is
6
uniformly distributed.
▶ Independence: Since
1 1 1
pX ,Y (x, y ) = = · = pX (x) pY (y ) for all x, y ,
36 6 6
X and Y are independent.
▶ Some probabilities:
 6 1
▶ P(X + Y = 7) = = (pairs:
36 6
(1,6),(2,5),(3,4),(4,3),(5,2),(6,1)).
6 1
▶ P(X + Y ≤ 4) = = (pairs:
36 6
(1,1),(1,2),(1,3),(2,1),(2,2),(3,1)).
P(X = 1, X + Y = 7) 1/36 1
▶ P(X = 1 | X +Y = 7) = = = .
P(X + Y = 7) 6/36 6
Expectation and Variance:
1+2+3+4+5+6
E [X ] = E [Y ] = = 3.5,
6
E [X + Y ] = E [X ] + E [Y ] = 7,
35
Var(X ) = Var(Y ) = ,
12
35
Var(X + Y ) = Var(X ) + Var(Y ) = (by independence).
6
Joint Probability Density Function (PDF)
For two continuous random variables X and Y :
Joint PDF
Z ∞ Z ∞
f (x, y ) ≥ 0 and f (x, y ) dx dy = 1
−∞ −∞

For any region A in the xy-plane:

ZZ
P[(X , Y ) ∈ A] = f (x, y ) dx dy
A

Example
For a uniform distribution over a unit square:

f (x, y ) = 1 for 0 ≤ x ≤ 1, 0 ≤ y ≤ 1
Example 3: Continuous Bivariate Distribution

Joint pdf:
(
2e −x e −y , x > 0, y > 0,
fX ,Y (x, y ) =
0, otherwise.

Marginal of X :
Z ∞ Z ∞
−x −y −x
fX (x) = 2e e dy = 2e e −y dy = 2e −x ·1 = 2e −x , x > 0.
0 0

By symmetry, fY (y ) = 2e −y , y > 0.
Example 3: Continuous Bivariate Distribution

Joint pdf:
(
2e −x e −y , x > 0, y > 0,
fX ,Y (x, y ) =
0, otherwise.

Marginal of X :
Z ∞ Z ∞
−x −y −x
fX (x) = 2e e dy = 2e e −y dy = 2e −x ·1 = 2e −x , x > 0.
0 0

By symmetry, fY (y ) = 2e −y , y > 0.
Independence check:

fX (x)fY (y ) = (2e −x )(2e −y ) = 4e −(x+y ) ̸= 2e −x e −y .

Thus X and Y are not independent.

Summary

▶ The joint distribution contains all information about the pair

(X , Y ).
▶ Marginal distributions are obtained by summing/integrating the
joint over the other variable(s).
▶ If the joint factorises into the product of the marginals, then X and
Y are independent.
▶ Examples illustrate the mechanics for both discrete and continuous
cases.
Conditional Probability and Independence

Conditional Distribution
f (x, y )
f (x|y ) = (continuous)
fY (y )
p(x, y )
p(x|y ) = (discrete)
pY (y )

Independence
X and Y are independent if and only if:

f (x, y ) = fX (x) · fY (y ) (continuous)

p(x, y ) = pX (x) · pY (y ) (discrete)

Solved Example 1

Problem: Roll a fair die twice. Let X = first roll, Y = second roll.
Solution:
1
p(x, y ) = ∀x, y ∈ {1, 2, . . . , 6}
36
1
P(X = 3, Y = 5) =
36
4 1
P(X ≤ 2, Y ≥ 5) = =
36 9
6
X 1 1
pX (x) = = for all x
36 6
y =1
Solved Example 2

Problem: Given f (x, y ) = 3x for 0 ≤ y ≤ x ≤ 1.

Marginals:
Z x
fX (x) = 3x dy = 3x 2 , 0 < x < 1
0
Z 1
3
fY (y ) = 3x dx = (1 − y 2 ), 0 < y < 1
y 2

Conditional:
3x 2x
f (x|y ) = 3
= , y ≤x ≤1
2 (1
2
−y ) 1 − y2
Solved Example 2 (Continued)

Compute P(X ≤ 12 , 14 < Y < 43 ):

Z 1/2 Z 1/2
P= 3x dx dy
y =1/4 x=y
Z 1/2 

1 3 2
= − y dy
1/4 4 2
1/2
y3

3 1
= y−
2 4 3 1/4
5
=
128
Independence check: fX (x)fY (y ) = 92 x 2 (1 − y 2 ) ̸= 3x = f (x, y )
Therefore, X and Y are not independent.
What is a Marginal Distribution?

Definition (Marginal Distribution)

The probability distribution of a single random variable obtained by
summing (discrete) or integrating (continuous) the joint
distribution over the other variable(s).

▶ Discrete:
X X
pX (x) = pX ,Y (x, y ), pY (y ) = pX ,Y (x, y ).
y x

▶ Continuous:
Z ∞ Z ∞
fX (x) = fX ,Y (x, y ) dy , fY (y ) = fX ,Y (x, y ) dx.
−∞ −∞
Example 1: Discrete Joint Table

Given the joint PMF:

Y =1 Y =2 Y =3
X =0 0.1 0.2 0.1
X =1 0.3 0.1 0.2

Marginal of X :

pX (0) = 0.1 + 0.2 + 0.1 = 0.4, pX (1) = 0.3 + 0.1 + 0.2 = 0.6.

Marginal of Y :

pY (1) = 0.1+0.3 = 0.4, pY (2) = 0.2+0.1 = 0.3, pY (3) = 0.1+0.2 = 0.3

P P
Check: pX = 1, pY = 1.
Example 2: Continuous Joint PDF
Joint PDF:
3
fX ,Y (x, y ) = (x 2 + y 2 ), 0 < x < 1, 0 < y < 1.
2
Marginal of X :
Z 1
3 2
fX (x) = (x + y 2 ) dy
0 2
1
y3

3 2
= x y+
2 3 y =0
 
3 1 3 1
= x2 + = x2 + , 0 < x < 1.
2 3 2 2
Marginal of Y (by symmetry):
1 3 2
fY (y ) = + y , 0 < y < 1.
2 2
R1 R1 3 2 1


Check: 0 fX (x) dx = 0 2 x + 2 dx = 1 (similarly for fY ).
Solved Example (Continued)

Verify sum = 1:

0.18 + 0.18 + 0.064 + 0.096 + 0.12 + 0.08 + 0.12 + 0.16 = 1.00

P(X = 0) = 0.18 + 0.18 + 0.064 + 0.096 = 0.52

P(X = 1, Y = 0) = 0.12 + 0.08 = 0.20

P(X = 0, Y = 1, Z = 1)
P(Z = 1|X = 0, Y = 1) =
P(X = 0, Y = 1)
0.096 0.096
= = = 0.6
0.064 + 0.096 0.16
What is Covariance?

Conceptual Definition
Covariance measures the linear relationship between two random
variables. It indicates how two variables change together.

Mathematical Definition:

Cov(X , Y ) = E [(X − µX )(Y − µY )]

where µX = E [X ] and µY = E [Y ]. Computational Formula:

Cov(X , Y ) = E [XY ] − E [X ]E [Y ]
Interpreting Covariance

▶ Cov(X , Y ) > 0: positive linear relationship

▶ Cov(X , Y ) < 0: negative linear relationship

▶ Cov(X , Y ) = 0: uncorrelated (no linear relationship)

Key Properties:
▶ Cov(X , Y ) = Cov(Y , X ) (symmetry)

▶ Cov(X , X ) = Var(X )

▶ Bilinearity: Cov(aX + b, cY + d) = ac Cov(X , Y )

▶ Var(X + Y ) = Var(X ) + Var(Y ) + 2 Cov(X , Y )

▶ Independence ⇒ zero covariance (converse not true)

Solved Example 1 (Discrete)

Joint PMF Table:

X \Y 0 1 2
0 0.1 0.2 0.1
1 0.2 0.1 0.1
2 0.0 0.1 0.1

Step 1: Marginals

pX (0) = 0.4, pX (1) = 0.4, pX (2) = 0.2

pY (0) = 0.3, pY (1) = 0.4, pY (2) = 0.3

Step 2: Means

E [X ] = 0.8, E [Y ] = 1.0
Covariance: Definition and Example
Definition
For two random variables X and Y :
 
Cov(X , Y ) = E (X − E [X ])(Y − E [Y ])

Cov(X , Y ) = E [XY ] − E [X ]E [Y ]

Simple Example: Joint distribution of X and Y :

Y =0 Y =1
X =0 0.2 0.3
X =1 0.1 0.4
P(X = 0) = 0.5, P(X = 1) = 0.5, P(Y = 0) = 0.3, P(Y = 1) = 0.7
E [X ] = 0 · 0.5 + 1 · 0.5 = 0.5, E [Y ] = 0 · 0.3 + 1 · 0.7 = 0.7.
Expected product E [XY ]
E [XY ] = (1 · 1) · 0.4 = 0.4.
Example (Continuous)
Joint PDF:

f (x, y ) = x + y , 0 ≤ x ≤ 1, 0 ≤ y ≤ 1

Step 1: Marginal Densities

Z 1
1
fX (x) = (x + y ) dy = x + , 0≤x ≤1
0 2
Z 1
1
fY (y ) = (x + y ) dx = y + , 0≤y ≤1
0 2
Step 2: Means
Z 1  
1 7
E [X ] = x x+ dx =
0 2 12
7
By symmetry, E [Y ] = 12 .
Example (Continued)
Step 3: Compute E [XY ]
Z 1Z 1 Z 1Z 1
E [XY ] = xy (x + y ) dy dx = (x 2 y + xy 2 ) dy dx
0 0 0 0

Inner integral:
1
x2 x
Z
(x 2 y + xy 2 ) dy = +
0 2 3
Outer integral:
Z 1 2 
x x 1 1 1
+ dx = + =
0 2 3 6 6 3

Step 4: Covariance
 2
1 7 1 49 1
Cov(X , Y ) = − = − =−
3 12 3 144 144

Weak negative linear relationship.

What is a Linear Combination?

A linear combination of random variables X1 , X2 , . . . , Xn is

Y = a1 X1 + a2 X2 + · · · + an Xn + b

where a1 , a2 , . . . , an and b are constants.

Examples:
▶ Portfolio return: P = 0.6A + 0.4B

▶ Difference of two measurements: D = X − Y

▶ Weighted sum: Z = 2X − 3Y + 1
Expected Value is Linear

For any constants ai and b:

E [Y ] = a1 E [X1 ] + a2 E [X2 ] + · · · + an E [Xn ] + b

Special Cases:

E [X + Y ] = E [X ] + E [Y ]
E [X − Y ] = E [X ] − E [Y ]
E [aX + b] = aE [X ] + b

This holds regardless of independence.

Variance Formula

n
X X
Var(Y ) = ai2 Var(Xi ) + 2 ai aj Cov(Xi , Xj )
i=1 1≤i<j≤n

If the variables are independent (Cov(Xi , Xj ) = 0 for i ̸= j):

Var(Y ) = a12 Var(X1 ) + a22 Var(X2 ) + · · · + an2 Var(Xn )

Important: Covariance terms can increase or decrease the

variance.
Solved Example

Problem: Portfolio: 60% in Stock A, 40% in Stock B.

E [A] = 8%, E [B] = 12%, Var(A) = 0.04, Var(B) = 0.09.
Independent returns. Solution:

P = 0.6A + 0.4B
E [P] = 0.6(8%) + 0.4(12%) = 4.8% + 4.8% = 9.6%
Var(P) = (0.6)2 (0.04) + (0.4)2 (0.09)
= 0.36 × 0.04 + 0.16 × 0.09 = 0.0144 + 0.0144 = 0.0288

Thus, expected return 9.6%, variance 0.0288.

Solved Example 2

Problem: E [X ] = 5, E [Y ] = 3, Var(X ) = 4, Var(Y ) = 9,

Cov(X , Y ) = 2. Find mean and variance of Z = 2X − 3Y + 1.
Solution:

E [Z ] = 2(5) − 3(3) + 1 = 10 − 9 + 1 = 2
Var(Z ) = 22 (4) + (−3)2 (9) + 2(2)(−3)(2)
= 4 × 4 + 9 × 9 + (−24) = 16 + 81 − 24 = 73

So E [Z ] = 2 and Var(Z ) = 73.

Definition of a Bernoulli Trial
Bernoulli Trial
A random experiment with exactly two outcomes:
▶ Success (S) with probability p

▶ Failure (F ) with probability 1 − p

The associated random variable X is:

(
1, success,
X =
0, failure.

Probability mass function:

P(X = 1) = p, P(X = 0) = 1 − p.

Key properties:

E [X ] = p, Var(X ) = p(1 − p).

Example: Free Throws
Example
A basketball player makes a free throw with probability 0.7
(success) and misses with probability 0.3 (failure). Assume each
free throw is independent.
(a) Define the Bernoulli random variable.
(
1, makes the shot,
X =
0, misses.

Here p = 0.7.
(b) Probability of success:

P(X = 1) = p = 0.7.

(c) Probability of failure:

P(X = 0) = 1 − p = 0.3.
Example: Free Throws (continued)

(d) Expected number of successes in one trial:

E [X ] = p = 0.7.

(e) Variance:

Var(X ) = p(1 − p) = 0.7 × 0.3 = 0.21.

Binomial Distribution

Definition
The binomial distribution counts how many times a specific
event (a “success”) happens when you repeat the same simple
experiment a fixed number of times, and each repetition is
independent and has the same chance of success.
Intuition: Flip a coin several times. The binomial distribution tells
you the probability of getting, say, exactly 3 heads in 5 flips.

Formula: For n trials, success probability p,

 
n k
P(X = k) = p (1 − p)n−k , k = 0, 1, . . . , n.
k
Example 1: Flipping a Fair Coin

Problem: Flip a fair coin 5 times. Probability of exactly 3 heads?

Solution:

n = 5, p = 0.5, k = 3
   
5 3 2 5
P(X = 3) = (0.5) (0.5) = (0.5)5
3 3
 
5 1
= 10, (0.5)5 =
3 32
1 10 5
P = 10 × = = = 0.3125
32 32 16
Thus, a 31.25% chance.
Example 2: Guessing on a Test

Problem: 10 questions, each with 4 choices. Random guessing.

Probability exactly 4 correct?
Solution:

n = 10, p = 0.25, k = 4
 
10
P(X = 4) = (0.25)4 (0.75)6
4
 
10
= 210, (0.25)4 = 0.00390625, (0.75)6 ≈ 0.17797
4
210 × 0.00390625 = 0.8203125
0.8203125 × 0.1779785 ≈ 0.146

About 14.6% chance.

Multinomial Distribution

Definition
The multinomial distribution is like the binomial, but instead of
just two outcomes (success/failure), there are several possible
outcomes. It gives the probability of getting specific counts for
each outcome in a fixed number of independent trials.

Intuition: Roll a die multiple times. The multinomial distribution

tells you the probability of getting, say, exactly two of each number
in 12 rolls.

Formula: For n trials, m categories with probabilities p1 , . . . , pm :

n!
P(X1 = x1 , . . . , Xm = xm ) = p x1 · · · pm
xm
.
x1 ! · · · xm ! 1
Example 1: Fair Die Rolled 6 Times

Problem: Roll a fair die 6 times. Probability each face appears

exactly once?
Solution:
1
n = 6, m = 6, pi = , xi = 1 for all i
6
 6
6! 1 1
P= = 720 ×
1!1!1!1!1!1! 6 46656
720 5
= ≈ 0.01543
46656 324
About 1.54% chance.
Example 2: Sampling Voters
Problem: 30% prefer A, 50% B, 20% C. Sample 5 people.
Probability exactly 2 A, 2 B, 1 C?
Solution:

n = 5, pA = 0.3, pB = 0.5, pC = 0.2

xA = 2, xB = 2, xC = 1
5!
P= × (0.3)2 (0.5)2 (0.2)1
2! 2! 1!
5! 120
= = 30
2!2!1! 4
(0.3)2 = 0.09, (0.5)2 = 0.25, (0.2)1 = 0.2
0.09 × 0.25 = 0.0225, 0.0225 × 0.2 = 0.0045
30 × 0.0045 = 0.135

Thus, 13.5% probability.

hypergeometric distribution

Definition
The hypergeometric distribution gives the probability of
obtaining exactly k successes in n draws without replacement
from a finite population of size N that contains exactly K
successes.
Intuition: Drawing cards from a deck without replacement. Each
draw changes the composition, so the probabilities are not
constant. Key difference from binomial: Binomial assumes
independent draws with replacement (constant probability);
hypergeometric assumes no replacement.
Mathematical Formula

K N−K



k n−k
P(X = k) = N


n
where:
▶ N = population size

▶ K = number of successes in the population

▶ n = sample size (drawn without replacement)

▶ k = number of successes in the sample

▶ ba = binomial coefficient


Example 1: Cards

Problem: From a standard deck (52 cards), draw 5 cards without

replacement. What is the probability of exactly 2 aces?
Solution:

N = 52, K = 4, n = 5, k=2
4 48



2
P= 52

3
5
     
4 48 52
= 6, = 17296, = 2598960
2 3 5
6 × 17296 103776
P= = ≈ 0.03993
2598960 2598960
About 3.99% chance.
Example 2: Defective Components

Problem: A batch of 20 components contains 3 defectives.

Randomly select 5 without replacement. Probability of exactly 1
defective?
Solution:

N = 20, K = 3, n = 5, k=1
3 17



1
P= 20

4
5
     
3 17 20
= 3, = 2380, = 15504
1 4 5
3 × 2380 7140
P= = ≈ 0.4605
15504 15504
About 46.05% chance.
Hypergeometric Distribution

Definition
The hypergeometric distribution gives the probability of
obtaining exactly k successes in n draws without replacement
from a finite population of size N that contains exactly K
successes.
Intuition: Drawing cards from a deck without replacement. Each
draw changes the composition, so the probabilities are not
constant. Key difference from binomial: Binomial assumes
independent draws with replacement (constant probability);
hypergeometric assumes no replacement.
Mathematical Formula

K N−K



k n−k
P(X = k) = N


n
where:
▶ N = population size

▶ K = number of successes in the population

▶ n = sample size (drawn without replacement)

▶ k = number of successes in the sample

▶ ba = binomial coefficient


Example 1: Cards

Problem: From a standard deck (52 cards), draw 5 cards without

replacement. What is the probability of exactly 2 aces?
Solution:

N = 52, K = 4, n = 5, k=2
4 48



2
P= 52

3
5
     
4 48 52
= 6, = 17296, = 2598960
2 3 5
6 × 17296 103776
P= = ≈ 0.03993
2598960 2598960
About 3.99% chance.
Example 2: Defective Components

Problem: A batch of 20 components contains 3 defectives.

Randomly select 5 without replacement. Probability of exactly 1
defective?
Solution:

N = 20, K = 3, n = 5, k=1
3 17



1
P= 20

4
5
     
3 17 20
= 3, = 2380, = 15504
1 4 5
3 × 2380 7140
P= = ≈ 0.4605
15504 15504
About 46.05% chance.
Negative Binomial Distribution

Definition
The negative binomial distribution models the number of trials
needed to achieve a fixed number of successes in a sequence of
independent and identical Bernoulli trials.

Intuition: Flip a coin until you get 3 heads. The number of flips
required (including the last head) is a negative binomial random
variable.

Key difference from binomial: Binomial fixes the number of

trials and counts successes; negative binomial fixes the number of
successes and counts trials.
Mathematical Formula

 
x −1 r
P(X = x) = p (1 − p)x−r , x = r , r + 1, r + 2, . . .
r −1

where:
▶ r = desired number of successes

▶ p = probability of success on each trial

▶ x = total number of trials (including the last success)

▶ x−1


r −1 counts ways to arrange the first r − 1 successes among the
first x − 1 trials.

r
Mean: E [X ] = p
r (1−p)
Variance: Var(X ) = p2
Example 1: Getting 3 Heads

Problem: A fair coin is flipped until 3 heads appear. What is the

probability that exactly 5 flips are needed?
Solution:

r = 3, p = 0.5, x = 5
   
5−1 4
P(X = 5) = (0.5)3 (0.5)2 = (0.5)5
3−1 2
 
4 1
= 6, (0.5)5 =
2 32
1 6 3
P =6× = = = 0.1875
32 32 16
There is an 18.75% chance.
Example 2: Making 2 Shots

Problem: A basketball player makes 70% of free throws. She

shoots until she makes 2. What is the probability she takes exactly
4 shots?
Solution:

r = 2, p = 0.7, x = 4
   
4−1 2 2 3
P(X = 4) = (0.7) (0.3) = (0.49)(0.09)
2−1 1
 
3
= 3, 0.49 × 0.09 = 0.0441
1
P = 3 × 0.0441 = 0.1323

About 13.23% chance.

Geometric Distribution?

Definition
The geometric distribution models the number of trials needed
to get the first success in a sequence of independent and identical
Bernoulli trials.

Intuition: Flip a coin until you get heads. The number of flips
required follows a geometric distribution.

Key properties:
▶ Trials are independent, each with success probability p.

▶ We stop at the first success.

▶ Possible values: k = 1, 2, 3, . . .
Mathematical Formula

P(X = k) = (1 − p)k−1 p, k = 1, 2, 3, . . .

1
Mean: E [X ] =
p
1−p
Variance: Var(X ) =
p2
Memoryless property: The geometric distribution is the only
discrete distribution that is memoryless:

P(X > m + n | X > m) = P(X > n)

Example 1: First Head on the 4th Flip

Problem: A fair coin is flipped until heads appears. Find the

probability that the first head occurs on the 4th flip.
Solution:

p = 0.5, k=4
P(X = 4) = (1 − p)4−1 p = (0.5)3 × 0.5
= 0.125 × 0.5 = 0.0625

Thus, there is a 6.25% chance.

Example 2: First 6 on the 3rd Roll

Problem: A fair die is rolled until a 6 appears. Find the

probability that it takes exactly 3 rolls.
Solution:
1 5
p = , 1−p = , k =3
6 6
 2
5 1 25 1 25
P(X = 3) = × = × = ≈ 0.1157
6 6 36 6 216

About 11.57% chance.

Poisson Distribution

Definition
The Poisson distribution models the number of times an event
occurs in a fixed interval of time or space, given that these events
happen with a known constant average rate and independently of
each other.

Intuition: Count emails per hour, calls per minute, or defects per
meter – all can be modeled with Poisson.

Key properties:
▶ Events occur independently.

▶ Average rate λ is constant.

▶ No two events occur simultaneously.

Mathematical Formula

e −λ λk
P(X = k) = , k = 0, 1, 2, . . .
k!

Mean: E [X ] = λ
Variance: Var(X ) = λ (equal mean and variance)

Conditions for Poisson:

▶ The probability of an event in a small interval is proportional to
the length.
▶ Events in disjoint intervals are independent.
Example 1: Calls per Minute

Problem: A call center receives 5 calls per minute on average.

Find the probability of exactly 3 calls in a minute.
Solution:

λ = 5, k=3
e −5 · 53 e −5 · 125
P(X = 3) = =
3! 6
Using e −5 ≈ 0.0067379:
0.0067379 × 125 0.8422375
P≈ = ≈ 0.14037
6 6
Thus, about 14.04% chance.
Example 2: Typos per Page

Problem: A book averages 2 typos per page. Find the probability

a given page has no typos.
Solution:

λ = 2, k=0
e −2 · 20
P(X = 0) = = e −2
0!
e −2 ≈ 0.135335, so about 13.53% chance.
Binomial Distribution
Binomial Distribution
X ∼ Binomial(n, p) counts the number of successes in n
independent Bernoulli trials, each with success probability p.
 
n k
P(X = k) = p (1 − p)n−k , k = 0, 1, . . . , n.
k

Mean and Variance

E [X ] = np , Var(X ) = np(1 − p) .

Example 1: Toss a fair coin 10 times. Let X = number of heads.

Here n = 10, p = 0.5. Then

E [X ] = 10 · 0.5 = 5, Var(X ) = 10 · 0.5 · 0.5 = 2.5.

P(X = 6) = 10 6 4 10 ≈ 0.205.


6 (0.5) (0.5) = 210 · (0.5)
Example 2: A multiple-choice test has 5 questions, each with 4
options. Random guessing gives p = 0.25, n = 5.

E [X ] = 5 · 0.25 = 1.25, Var(X ) = 5 · 0.25 · 0.75 = 0.9375.

P(X = 2) = 52 (0.25)2 (0.75)3 = 10 · 0.0625 · 0.421875 ≈ 0.2637.

Poisson Distribution
Definition
X ∼ Poisson(λ) models the number of events occurring in a fixed
interval, with average rate λ > 0.

λk
P(X = k) = e −λ , k = 0, 1, 2, . . .
k!

Mean and Variance

E [X ] = λ , Var(X ) = λ .

Example 1: Emails arrive at an average rate of 5 per hour. Let X

= number of emails in one hour, λ = 5.
53 125
E [X ] = 5, Var(X ) = 5, P(X = 3) = e −5 = e −5 ≈ 0.1404.
3! 6
Example 2: A book has an average of 0.2 typos per page. For one
page, λ = 0.2.
Geometric Distribution (Trials until first success)
Definition
X ∼ Geometric(p) counts the number of independent Bernoulli
trials needed to obtain the first success, where each trial has
success probability p.

P(X = k) = (1 − p)k−1 p, k = 1, 2, 3, . . .

Mean and Variance

1 1−p
E [X ] = , Var(X ) = .
p p2

Example 1: Roll a fair die until a 6 appears. p = 16 .

1 5/6 5/6
E [X ] = = 6, Var(X ) = 2
= = 30.
1/6 (1/6) 1/36
P(first 6 on 3rd roll) = (1 − 16 )2 · 16 = ( 65 )2 · 16 = 216
25
≈ 0.1157.
Example 2: A shooter hits the target with probability 0.8 per
Hypergeometric Distribution
Definition
X ∼ Hypergeometric(N, K , n) counts the number of successes in a
sample of size n drawn without replacement from a population of
size N containing K successes.
K N−K



k n−k
P(X = k) = N

, max(0, n − (N − K )) ≤ k ≤ min(n, K ).
n

Mean and Variance

 
K K K N −n
E [X ] = n , Var(X ) = n 1− .
N N N N −1

Example 1: Draw 5 cards from a standard deck (52 cards, 4

aces). N = 52, K = 4, n = 5.
4 20 4 48 47 5 · 4 · 48 · 47
E [X ] = 5· = ≈ 0.3846, Var(X ) = 5· · · = ≈
52 52 52 52 51 52 · 52 · 51
Negative Binomial Distribution (Trials until r -th success)
Definition
X ∼ Negative Binomial(r , p) counts the number of independent
Bernoulli trials needed to obtain r successes, each with success
probability p.
 
k −1 r
P(X = k) = p (1 − p)k−r , k = r , r + 1, . . .
r −1

Mean and Variance

r r (1 − p)
E [X ] = , Var(X ) = .
p p2

Example 1: Roll a fair die until the 3rd time a 6 appears. r = 3,

p = 16 .
3 3 · (5/6) 5
E [X ] = = 18, Var(X ) = 2
= 3 · · 36 = 90.
1/6 (1/6) 6
Frame Title

END OF LECTURE