Mean and variance, and link to Poisson processes
Studying this course? Keep your progress in one place.
Key facts
- , and .
- The proofs use integration by parts; state that the boundary term is .
- In a Poisson process, counts in time are and gaps are exponential with parameter .
- .
Mean, variance and standard deviation
The mean and standard deviation are both ; the variance is .
For exponential with parameter :
The mean and standard deviation are equal. A higher rate means shorter gaps on average.
- Mean
- Variance
- Standard deviation
Worked example
per day. Find the mean, variance and standard deviation.
- 1
Mean days.
- 2
Variance .
- 3
Standard deviation days.
An exponential distribution has mean 4. What is its variance?
Proofs of the mean and variance
Integrate by parts; the boundary terms vanish because the exponential decays faster than grows.
Mean. . By integration by parts with and :
Variance. By parts again, .
State the boundary term is zero because decays faster than grows.
- 1
Integrate by parts
u = x, v = −e^(−λx)
- 2
Boundary term
zero at both ends
- 3
Remaining integral
gives 1/λ
- 4
E(X²)
by parts again, reusing E(X)
- 5
Subtract
Var = 2/λ² − 1/λ² = 1/λ²
Which expression gives for an exponential distribution?
The link to Poisson processes
In a Poisson process counts are Poisson and gaps are exponential, with the same rate .
Events occurring at random, independently, at a constant average rate per unit time form a Poisson process. Then:
- the number of events in a time is Poisson with mean ,
- the time between successive events is exponential with parameter .
The mean gap is . For example 12 strikes per hour means per minute, so the mean gap is 5 minutes.
| How many events in a period t? | How long until the next event? | |
|---|---|---|
| Question | Counting events | Measuring time |
| Distribution | Poisson, mean | Exponential, parameter |
How many events in a period t?
- Question:
- Counting events
- Distribution:
- Poisson, mean
How long until the next event?
- Question:
- Measuring time
- Distribution:
- Exponential, parameter
Faults occur at a rate of 6 per hour. What is the mean time between faults?
Linking the two distributions
A gap longer than is the same event as no events in time , with probability .
A gap longer than is the same event as no events in a time . Both have probability :
Use the Poisson count for "how many events in a period", and the exponential for "how long until the next event".
If the rate varies, for instance faults are more common near joints, neither model applies.
- Gap longer than no events in time
- Probability
Worked example
Faults occur at per km. Find the probability that the gap between faults is longer than 2 km.
- 1
A gap longer than 2 km means no faults in 2 km.
- 2
The number of faults in 2 km is Poisson with mean .
- 3
Events form a Poisson process with rate . What is the probability of no events in time ?
Try an exam question
Faults occur randomly along a cable at a constant average rate of per km. Find the probability that the distance between two successive faults is more than km.
[3 marks]
- [1]A gap longer than 2 km means no faults in 2 km.
- [1]The number of faults in 2 km is Poisson with mean .
- [1].
That's the notes covered.
Carry on to the next subtopic.
Try a question
Find the mean lifetime of a battery.
1 mark, marked against the mark scheme
Find the mean lifetime of a battery.
1 mark, marked against the mark scheme