Scribd
Upload
69 views

Stochastic Models: Stochastic Processes: Bernoulli Process, Poisson Process)

This document summarizes key stochastic processes: the Bernoulli process and Poisson process. The Bernoulli process models events over discrete time that occur with probability ρ and not with probability 1-ρ. The interarrival times follow a geometric distribution. The Poisson process generalizes this to continuous time, where the interarrival times follow an exponential distribution. The number of arrivals in a time period t follows a Poisson distribution with rate λ and mean λt. Both processes have the memoryless property.

Uploaded by

mod nod
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
69 views

Stochastic Models: Stochastic Processes: Bernoulli Process, Poisson Process)

This document summarizes key stochastic processes: the Bernoulli process and Poisson process. The Bernoulli process models events over discrete time that occur with probability ρ and not with probability 1-ρ. The interarrival times follow a geometric distribution. The Poisson process generalizes this to continuous time, where the interarrival times follow an exponential distribution. The number of arrivals in a time period t follows a Poisson distribution with rate λ and mean λt. Both processes have the memoryless property.

Uploaded by

mod nod
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Stochastic Models

Lecture 2
(Stochastic Processes:
Bernoulli Process, Poisson Process)

by
Dr. Olivia Morad
2018
1
Contents:
1- Bernoulli Process
• Bernoulli Distribution and Geometric Distribution

2- Poisson Process
• Poisson distribution and Exponential distribution

2
1- Bernoulli Process

Definition:

• Bernoulli Process is a finite or infinite sequence of binary random


variables, so it is a discrete time stochastic process that takes only two
values 0, 1. (Time slot not time interval).

• Bernoulli process is a sequence Y1,Y2, ….. of IID binary random


variables, (Independent Random Variables and have the same
probability distribution (Bernoulli distribution).

3
• Bernoulli Distribution:
•Parameters 0<ρ<1 ρ∈R

• Pmf ρ for k = 1 success

q = (1 - ρ) for k = 0 failure
• Mean ρ
• Variance ρ (1 – q)

• Let ρ = Pr {Yi = 1} and 1- ρ = Pr {Yi = 0}

• We usually visualize Bernoulli process as evolving in discrete time with the


event {Yi = 1} → representing an arrival of customer at time i and {Yi = 0}
→ no arrival.
4
• Discrete-time, Discrete state space.

• Actually the Bernoulli Process could be characterized by a sequence of


interarrival time not only a sequence of arrivals (i,e, IID a binary variable of
arrivals {0, 1}). These interarrival times are IID r.v geometrically distributed
X1, X2,…

• Memoryless (The time to the next arrival is independent of the past).

5
• Geometric Distribution:
•Parameters 0<ρ≤1 success probability

• Pmf Pr (X = k) = (1 - ρ) k-1 ρ, k = 1, 2, 3 ……..

• The probability distribution of the number X of Bernoulli trials needed to


get one success.

• Mean 1
ρ
1− ρ
• Variance ρ2

6
2- Poisson Process
• Simple/widely used stochastic process to model the times at which
arrivals enter a system.

• Continuous time version of the Bernoulli process.


⇒ no time slot but time intervals.

• The intervals lengths may not be equals.

7
• Poisson process arrival may occur at arbitrary position of times and the
probability of arrival at any instance is 0.
⇒Define the process in terms of sequence of interarrival times.

• The exponential distribution is fundamental in the theory of continuous-


time Markov chains, due in major part to its memoryless property.

• A Poisson process is a stochastic process in which the interarrival times


have an exponential distribution function: i,e, for some real λ > 0, each Xi
has the density
fX(x) = λexp (- λx) for x ≥ 0

8
* Distribution of number of arrivals during a period of time:
Poisson distribution.

• Parameters: λ > 0
• arrival rate E [# of arrivals]/ per unit of time
• Pmf: PN(t ) (n) =
(λt ) exp(− λt )
n

n!

The probability of having n number of arrivals in (0, t])


• Mean: λt
• Variance: λt

9
* Distribution interarrival times: "Exponential
distribution" continuous analog of to the geometric
distribution

• Parameter λ>0
• Pdf λ e-λx
• Mean λ-1
• Variance λ-2
• Continuous time, Discrete state space.
• Interarrival times (IID), exponentially distributed.
• Memory less property.

10
THANK YOU

11

You might also like