MATH2209 Chap.3 Binomial and Poisson Models
Binomial and Poisson Models
Both distributions describe counts, but their data-generating mechanisms are different. The binomial model counts successes across a fixed number of independent binary trials with constant success probability. The Poisson model counts events over a stated exposure at a stable rate.
This chapter turns those conditions into a model-selection checklist, then works exact probabilities, expected counts and spread while keeping boundary words such as exactly, at most and at least explicit.
What this chapter covers
- 01
Binomial trial conditions
- 02
Exact and cumulative counts
- 03
Binomial mean and spread
- 04
Poisson rates and exposure
- 05
Poisson probability calculations
- 06
Boundary language and tail checks
Worked example · free
Count four detections in five assays
- 1The setting is binomial with n = 5 and p = 0.80.
- 1Exactly four uses C(5,4)(0.80)^4(0.20).
- 1C(5,4) = 5 and (0.80)^4 = 0.4096.
- 1The probability is 5 x 0.4096 x 0.20 = 0.4096.
Key terms
- Binomial Trial
- One independent binary trial with the same success probability as the others.
- Poisson Rate
- The expected event count per stated unit of exposure.
- Expected Count
- The long-run average value of a count random variable.
Binomial and Poisson Models FAQ
How do I distinguish binomial from Poisson?
Look for the mechanism. Binomial questions fix a number of binary trials and count successes. Poisson questions fix an exposure interval and count events arising at a rate. A count alone does not determine the model.
Why must a Poisson rate be rescaled?
The parameter is rate multiplied by exposure. A rate of three per hour implies an expected count of six over two hours and 1.5 over half an hour. Using the unadjusted rate answers a different exposure question.
What is the difference between at least and exactly?
Exactly four includes only X = 4. At least four includes X = 4, 5 and any larger supported counts. At most four includes all supported values from zero through four. Writing the included values prevents cumulative-tail mistakes.
How can I verify a discrete-model answer?
Write the support and expected value before judging the output. A binomial count must lie from zero to the fixed n, while a Poisson count can be any nonnegative integer. Check that binomial conditions identify fixed trials, binary outcomes, constant probability and independence; for Poisson, check that the rate and exposure use compatible units.
Compare the requested count with the mean to see whether a large or small probability is plausible, and ensure an exact event was not replaced by a cumulative one. Retain the trial set or exposure interval in the conclusion. If software is used, list the included counts or shaded tail first so that the chosen cumulative option can be audited rather than trusted from its label alone.
Assessment move
Create short scenarios and identify the four binomial conditions or the Poisson rate and exposure. For a binomial candidate, write fixed n, two outcomes, constant p and independence beside the story. If any condition is missing, say what extra assumption would be needed rather than applying the formula automatically. Define the random variable in a sentence and draw its support from zero to n.
Circle the values represented by exactly, at most, fewer than and at least before opening software; boundary translation is a separate skill from calculation. Recompute one exact probability by hand with the combination term, the success power and the failure power on separate lines. Check that the mean np lies inside the support and that the standard deviation uses np(1-p), not np alone.
For Poisson work, attach units to the rate and exposure. Multiply them only after confirming compatible units, then state that lambda-times-t is both the expected count and the variance for the model. Practise changing a rate per hour into a two-hour or half-hour mean. Keep the distinction between an underlying rate and the expected count over the chosen interval visible in the notation.
Compare exact and cumulative questions by listing the included counts. After every numerical result, locate the requested count relative to the model mean and decide whether the probability magnitude is plausible. Counts far into a tail should not accidentally receive the largest probability. Build mixed sets in which the wording alone does not announce the model.
Decide from the mechanism: fixed binary trials suggest binomial; events arising in exposure suggest Poisson. Include examples where neither model is justified, such as changing success probabilities or clustered events, and explain the failure. Finish by writing a one-sentence interpretation with the count, interval or trial set, and model assumptions.
This prevents a correct decimal from being detached from the experiment it describes.
Working through Binomial and Poisson Models in MATH2209? Sia is AskSia’s AI Statistics tutor — ask any MATH2209 Binomial and Poisson Models question and get a clear, step-by-step explanation grounded in how MATH2209 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.