Scribd
Upload
81 views7 pages

The Geometric Distribution: Univariate Distributions

The document discusses the geometric distribution as a discrete probability distribution that describes the number of trials needed for the first success in a sequence of independent Bernoulli trials. It defines the geometric distribution with parameter p as the probability of success on each trial. Key properties discussed include the memoryless property, the probability mass function p(x) = p(1-p)^(x-1), the expected value being 1/p, and the variance being (1-p)/p^2. Examples of geometric distributions with different p values are also shown.

Uploaded by

big clap
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
81 views7 pages

The Geometric Distribution: Univariate Distributions

The document discusses the geometric distribution as a discrete probability distribution that describes the number of trials needed for the first success in a sequence of independent Bernoulli trials. It defines the geometric distribution with parameter p as the probability of success on each trial. Key properties discussed include the memoryless property, the probability mass function p(x) = p(1-p)^(x-1), the expected value being 1/p, and the variance being (1-p)/p^2. Examples of geometric distributions with different p values are also shown.

Uploaded by

big clap
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

Univariate Distributions

The Geometric Distribution

Univariate Distributions

School of Risk & Actuarial Studies


UNSW Business School

Video lecture notes

1/7
Univariate Distributions
Geometric Distribution

The Geometric Distribution

I Discrete distribution.
I Consider a sequence of independent and repeated trials with p
denoting the probability of success.

I Suppose X is the number of trials required to obtain a first


success. Then X has a geometric distribution and we write
X ∼ Geometric (p) where 0 < p ≤ 1 denotes the probability of
a success for one trial.

2/7
Univariate Distributions
Geometric Distribution

The Geometric Distribution

I Application: the time (in years) to the first earthquake in


Australia with a magnitude larger than 8.

I Note: due to i.i.d. trials: the Geometric distribution has the


memoryless property, i.e., Pr(X ≤ a|X ≥ b) = Pr(X ≤ a − b)
for b < a.

I Exercise: Prove an equivalent statement of the memoryless


property is Pr[X > a + b | X > b] = Pr[X > a] for
a > 0, b > 0, both positive integers.

3/7
Univariate Distributions
Geometric Distribution

The Geometric Distribution

I Probability mass function:

pX (x) = p · (1 − p)x−1 (1)

for x = 1, 2, . . .. parameter constraints: 0 < p ≤ 1.


I Expected value (mean):
X
E [X ] = x · p · (1 − p)x−1
all x
∞
X
=p · x · (1 − p)x−1
x=1

4/7
Univariate Distributions
Geometric Distribution

The Geometric Distribution


I Using geometric series: for |a| < 1,
∞ ∞
X 1 X 1 a
ax = ⇔ ax = −1= . (2)
1−a 1−a 1−a
x=0 x=1

Differentiating both sides with respect to a,


∞
X 1
xax−1 = . (3)
(1 − a)2
x=1

I Expected value (mean):

1 1
E [X ] = p · 2
= .
p p
5/7
Univariate Distributions
Geometric Distribution

The Geometric Distribution


I Variance: Var (X ) = 1−p
p2
(similar as mean).
I Moment generating function:
h i X∞
MX (t) =E e X ·t = e x·t · p · (1 − p)x−1
x=1
∞
X
t
=p · e · e t·z · (1 − p)z
z=0
∞
X z
=p · e t · e t · (1 − p)
z=0
p · et
= .
1 − (1 − p) · e t
 
1
Note: (1 − p) · e t < 1 ⇒ t < log 1−p .
6/7
Univariate Distributions
Geometric Distribution

Example of the Geometric Distribution

Geo(0.5) p.m.f. Geo(0.1) p.m.f.


probability mass function

probability mass function


p = 0.5 p = 0.1
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
0 5 10 15 0 10 20 30 40 50
x x
Geo(0.5) c.d.f. Geo(0.1) c.d.f.
cumulative density function

cumulative density function


1 1

0.5 0.5

p = 0.5 p = 0.1
0 0
0 5 10 15 0 10 20 30 40 50
x x

7/7

You might also like