Hypothesis test for the mean of a Poisson distribution
Studying this course? Keep your progress in one place.
Key facts
- The null hypothesis and the alternative hypothesis are about the parameter , never about .
- Scale to the length of the observation period before you start.
- Under , . Use for an increase and for a decrease.
- For a two-tailed test put half the significance level in each tail.
- Conclude in context: reject , or there is insufficient evidence; never "proved".
The Poisson setting
A single observed count is used to decide whether the mean rate has changed from its stated value.
The Poisson distribution models the number of events in a fixed interval of time or space. It needs events that occur randomly, independently and at a constant average rate. The parameter is the mean number of events in the interval.
A hypothesis test for the mean uses a single observed count to decide whether the population parameter has changed from its stated value.
If the sample covers several intervals, scale the mean first. A rate of errors per page gives for pages.
- Rate per interval, intervals
Calls arrive at a mean rate of per hour. What is the distribution for a -hour period?
Stating the hypotheses
Write and an alternative in terms of the parameter .
The null hypothesis says the parameter has its stated value: . The alternative hypothesis says how it differs. Write both in terms of , not the random variable or the observed value.
Define in context, for example "the mean number of calls per hour". Then state the distribution of the test statistic under , for instance .
Writing is a common error: hypotheses are about the parameter.
One-tailed test
- Increase:
- Decrease:
Two-tailed test
- Change only:
A shop usually gets 5 complaints a day. A manager thinks the mean has fallen. Which alternative hypothesis is correct?
One-tailed tests
Find the probability of a result at least as extreme as the observation and compare it with the significance level.
Assume is true, so . Find the probability of a result at least as extreme as the one observed and compare it with the significance level. If it is smaller, reject .
The critical region is the set of values that leads to rejecting . For and at : but , so the critical region is . The actual significance level is .
Worked example
Calls arrive at a mean rate of per hour. In one hour calls are received. Test at the level whether the rate has increased.
- 1
, . Under , .
- 2
Find .
- 3
Compare: , so reject .
For , . Is in the critical region for at ?
Two-tailed tests
Split the significance level equally between the two tails.
For , split the significance level equally between the two tails. At use in each tail: the lower tail and the upper tail .
For an observed value in the upper tail, compare with (or double it and compare with ).
Two-tailed critical regions have two parts. Give both, and add the two tail probabilities to get the actual significance level.
- Significance level , two-tailed in each tail
Worked example
under and at the level. Find the critical region and the actual significance level.
- 1
Lower tail: but , so .
- 2
Upper tail: but , so .
- 3
Add the two tail probabilities: .
A two-tailed test is carried out at the 5% level. What probability goes in each tail?
Writing the conclusion
State what happens to , then write a sentence in context.
Finish in two steps. First state what happens to ("reject " or "do not reject "). Then write a sentence in context, such as "There is sufficient evidence, at the level, that the mean number of calls per hour has increased."
If the result is not significant, say there is insufficient evidence that the rate has changed, rather than stating that the rate is unchanged. A non-significant result does not prove .
If asked to comment on the model: bursts or a varying rate break the Poisson conditions, so counts vary more than expected and the test may be unreliable.
- 1
Hypotheses
State and using , defined in context
- 2
Distribution
Write under
- 3
Probability
Find the tail probability or critical region
- 4
Compare
Compare with the significance level
- 5
Conclude
Reject or do not reject, then a sentence in context
The p-value is 0.12 at the 5% level. Which conclusion is best?
Try an exam question
Calls to a helpline arrive at a mean rate of per hour. In one hour calls are received. Test, at the significance level, whether the rate of calls has increased.
[5 marks]
- [1], .
- [1]Under , .
- [1].
- [1].
- [1]Reject : there is sufficient evidence, at the level, that the mean number of calls per hour has increased.
That's the notes covered.
Carry on to the next subtopic.
Try a question
Which of the following gives the correct hypotheses, where is the mean number of calls per hour?
1 mark, marked against the mark scheme
Which of the following gives the correct hypotheses, where is the mean number of calls per hour?
1 mark, marked against the mark scheme