Scribd
Upload
11 views14 pages

Stochastic Processes Presentation Group 8

Uploaded by

imahmoodaslam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
11 views14 pages

Stochastic Processes Presentation Group 8

Uploaded by

imahmoodaslam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

Topic: Poisson & Birth-death

Processes
Course: Stochastic Processes
Poisson Process
A Poisson process is a model for a series of discrete events where the
average time between events is known, but the exact timing is
random. The arrival of an event is independent of the event before.

The following properties define a Poisson process:

1.Events occur one at a time: There is either zero or one event in


each infinitesimal interval.
2.The occurrence of events in disjoint time intervals is
independent: Events in different, non-overlapping intervals happen
independently.
3.Stationary increments: The probability of a given number of
events occurring in a time interval only depends on the length of the
interval, not on its position in time.
1.Events occur one at a time: If you're counting people walking by,
you can imagine that in any tiny fraction of a second, either one
person walks by or no one does.
2.The occurrence of events in disjoint time intervals is
independent: The number of people who walk by in one period of
time doesn't affect the number of people who walk by in another
period. For example, how many people walk by between 2 PM and 3
PM is independent of how many people walked by between 1 PM and
2 PM.
3.Stationary increments: The likelihood of people walking by in a
given amount of time only depends on the length of that time period,
not on when it happens. So, the probability of seeing 5 people in an
hour is the same whether it's 9 AM to 10 AM or 3 PM to 4 PM.
How It Works

1.Rate (λ): This is the average number of events (like people


walking by) per unit of time (like per hour). If the rate λ is 5, it
means you expect, on average, 5 people to walk by per hour.
2.Counting Events: The Poisson process helps you figure out the
probability of different numbers of people walking by in a given
period. For example, you can calculate the chance that exactly 3
people walk by in 2 hours.
Derivation of the Poisson Process Formula
Notation
•𝜆: Average rate of events per unit time.
•𝑁(𝑡): Number of events up to time 𝑡.

Derivation Steps
1. Small Interval Probability:
1. Consider a very small interval of time Δ𝑡.
2. The probability of one event occurring in this interval is
approximately 𝑡 Δ𝑡.
3. The probability of no events occurring in this interval is
approximately 1−𝑡 Δ𝑡.
4. The probability of more than one event occurring is negligible
for very small Δ𝑡.
2. Divide Time into Small Intervals:
1. Divide the time period 𝑡t into 𝑛n small intervals of length Δ𝑡=𝑡𝑛.
2. As 𝑛 grows, Δ𝑡 becomes very small.
3. Probability of k Events:
1. The event count 𝑁(𝑡) in 𝑡 can be seen as the sum of the number of
events in each of the 𝑛 small intervals.
2. If 𝑘 events occur in time 𝑡, they can happen in (𝑛𝑘) ways
(combinations of choosing 𝑘 intervals out of 𝑛).
4. Calculate the Probability:
1. The probability of exactly 𝑘 events in 𝑡 is:𝑃(𝑁(𝑡)=𝑘)=(𝑛𝑘)
(𝜆Δ𝑡)𝑘(1−Δ𝜆𝑡) 𝑛−𝑘
2. As 𝑛→∞, Δ𝑡→0.
5.Using Limits:
•Using the binomial theorem and properties of limits:
(1-λΔt)(n-k) ≈ e-λt
•The term (nk)(λΔt)k simplifies to (λt)k/k!
6.Poisson Distribution:
•Combining these, the probability 𝑃(𝑁(𝑡)=𝑘) converges to:
Summary of the Poisson Formula
The Poisson process describes the probability of k events occurring in a
fixed time period t with a given average rate λ.

P(𝑁(𝑡)=𝑘) = (λtk * e-λt)/k!

This formula tells us the probability of observing exactly k events in


time t, where:
λt is the expected number of events.
e-λt accounts for the decreasing probability with increasing t.
λtk/k! distributes this probability among different counts of events k.
Examples

1.Inter-arrival Times: The time between one person walking by


and the next is random, but on average, it takes a certain amount of
time that depends on the rate. If the rate is 5 people per hour, then
on average, you see a person every 12 minutes.
2.Probability Calculation: If you want to know the probability of
seeing exactly 3 people walk by in 2 hours when the rate is 5
people per hour, you use the Poisson formula:
3.𝑃(3 people in 2 hours)= ( e-(5*2) * (5*2)3) / 3!​
This calculates the likelihood of that specific number of events
happening in that time period.
Birth Death Processes
A birth-death process is a type of stochastic process used in probability
theory and stochastic modeling to describe systems where individuals
or entities can "birth" or "die" with certain probabilities. It's often used
to model systems where entities enter and exit in a structured or
sequential manner, such as population dynamics, queuing systems, or
chemical reactions.
Here are the key components of a birth-death process:

1.State Space: The system can be in different states, often


represented by non-negative integers (0, 1, 2, ...). Each state
represents the number of entities present in the system.
2.Birth and Death Rates: At each state i, entities can either
give birth or die. The rates at which these events occur depend
on the current state of the system. Let λi be the birth rate and
μi​be the death rate when the system is in state i.
3. Transitions: Entities can move between states due to birth
or death events. If the system is in state i, it can transition to
state i+1 with rate λi (birth) or to state i−1 with rate μi
(death).
4. Markov Property: Like many stochastic processes, the
birth-death process typically satisfies the Markov property,
meaning that the system’s future behavior depends only on
its current state, not on how it arrived at that state.
Let’s denote Pi(t) as the probability that the system is in state i at time t.
The birth-death process can be described by the following set of
differential equations, known as the Kolmogorov forward equations:

Where:
•λi is the birth rate when the system is in state i.
•μi is the death rate when the system is in state i.
These equations represent the rate of change of probability for being in
state i at time t. The first term on the right-hand side represents the rate at
which the system transitions from state i−1 to state i due to birth events,
while the second term represents the rate of transition from state i+1 to
state i due to death events. The last term represents the rate at which the
system stays in state i due to either birth or death events.

These equations form a system of coupled ordinary differential equations


(ODEs) that can be solved numerically or analytically under specific
conditions. The solutions to these equations provide the time evolution of
the probabilities Pi(t), describing how the system's state probabilities
change over time in the birth-death process.

You might also like