University of Sydney · FACULTY OF STATISTICS

BUSS1020 Chap.4 Discrete Probability Distributions

- one subject, every graph, every model, every mark
6 Chapters3-page Bible
Our own words - no uploaded lecturer files
Updated for this semester
Chapter 4 of 10 · BUSS1020

Discrete Probability Distributions

Discrete distributions model countable outcomes. This chapter starts with probability distributions, expected value and variance, then separates three common processes. A binomial model needs fixed independent trials with two outcomes and constant probability. A Poisson model counts events over a stated exposure with a stable rate. A hypergeometric model handles finite sampling without replacement.

The examples make the model conditions visible before evaluating probabilities and interpret expected values over repeated processes. The same quality story is deliberately reframed under fixed trials, rate exposure and finite sampling so model selection follows the process rather than the word count. Expected value is treated as a long-run centre, not a possible fractional outcome for one batch.

Variance and standard deviation support reasonableness checks, while stable-rate and independence assumptions are challenged with campaign shocks, batch effects and time-of-day changes. Both practice calculations finish with an exposure or population-composition statement, which prevents a bare probability from being detached from the process that generated it.

A centre-based check also catches implausible outputs before they reach a decision. Keep every parameter attached to its trial count, exposure unit or finite population before comparing model outputs.

In this chapter

What this chapter covers

  • 01

    Discrete random variables and probability mass

  • 02

    Expected value and variance

  • 03

    Binomial conditions and probabilities

  • 04

    Poisson rates and exposure

  • 05

    Hypergeometric sampling without replacement

  • 06

    Covariance and sums of random variables

Worked example · free

Count successful pilot campaigns

Q [5 marks]. Eight independent pilot campaigns each have probability 0.30 of exceeding a target. Find the probability of exactly three successes, the expected count and the standard deviation. This is an AskSia-authored drill; the mark allocation shown here is illustrative and not an official university assessment scheme.
The first modelling question is structural: fixed independent trials suggest binomial, event counts over an interval suggest Poisson, and finite sampling withouThe first modelling question is structural: fixed independent trials suggest binomial, event counts over an interval suggest Poisson, and finite sampling without replacement suggests hypergeometric.count variablebinomialfixed trialsPoissonrate intervalhypergeometricno replacement
The first modelling question is structural: fixed independent trials suggest binomial, event counts over an interval suggest Poisson, and finite sampling without replacement suggests hypergeometric.
  • 1State X as Binomial with n equal to eight and p equal to 0.30.
  • 1Count the placements of three successes with C(8,3), which equals 56.
  • 1Evaluate 56 times 0.30 cubed times 0.70 to the fifth, approximately 0.2541.
  • 1Compute expected count np, which equals 2.4.
  • 1Compute standard deviation as the square root of np(1-p), approximately 1.296.
The probability of exactly three successes is about 0.2541, the expected count is 2.4 and the standard deviation is about 1.296 campaigns.
Sia tip — Prove the four binomial conditions before using the formula; the calculator cannot repair a mismatched count process.
Glossary

Key terms

Probability mass function
A rule assigning probabilities to the possible values of a discrete random variable.
Expected value
The probability-weighted long-run average over repeated realisations of a random variable.
Binomial distribution
A model for successes across fixed independent trials with constant success probability.
Poisson distribution
A model for event counts across a stated exposure under a stable-rate process.
Hypergeometric distribution
A model for successes drawn without replacement from a finite population with known composition.
FAQ

Discrete Probability Distributions FAQ

How do I choose between binomial and Poisson?

Choose binomial for a fixed number of independent two-outcome trials with constant success probability. Choose Poisson for a count of events over a time, area or other exposure with a stable rate. The random process matters more than similar-looking counts.

Why does without replacement change the model?

Without replacement, each draw changes both the remaining population size and possibly the remaining success count. The success probability is therefore not constant. A hypergeometric model tracks that depletion exactly for a finite population.

Can an expected count be non-integer?

Yes. Expected value describes the average over many repetitions, not a prediction that one realised count will equal it. A binomial expectation of 2.4 means repeated groups would average about 2.4 successes, even though each group has an integer count.

Study strategy

Exam move

Build a model-selection table with columns for random quantity, fixed trials, replacement, rate exposure and parameter. For each question, circle the phrase that identifies the process before calculating. Rehearse one binomial probability by expanding the combination and outcome probabilities. Scale a Poisson rate into the requested interval with units.

For hypergeometric questions, write the population success and failure counts above the combinations. Interpret expected value over repetitions and state when independence or a stable rate is doubtful. Compare three versions of one quality-control story: a fixed number of independent checks, a count of faults per metre and a sample drawn without replacement from a known shipment.

Explain why the same word success can lead to three models. For binomial questions, list n, p, x and the four conditions before using software. For Poisson questions, attach an exposure unit to lambda and rescale it when the time or area changes. For hypergeometric questions, verify that chosen success and failure counts add to the selected sample and that population counts deplete.

Use expected value as a repeated-process benchmark, then compare the requested count with that centre before judging whether the probability is reasonable. Inspect dependence from shared campaigns, batch effects or changing rates rather than repeating the word independent without evidence.

Finally, rewrite the interpretation so it states a probability under a model and does not promise the next batch will equal the expected count.

Working through Discrete Probability Distributions in BUSS1020? Sia is AskSia’s AI Statistics tutor — ask any BUSS1020 Discrete Probability Distributions question and get a clear, step-by-step explanation grounded in how BUSS1020 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.

A+Everything unlocked
Unlocks this Bible + all 202 of your subjects at University of Sydney, plus 1,000+ more Bibles across the AskSia library.
Sia - your BUSS1020 tutor, unlimited, worked the way the exam marks it
The full 3-page Bible + practice bank with worked solutions
Chrome extension - sync your LMS so Sia knows your deadlines
Bilingual EN / Chinese on every Bible and every Sia answer
$0.99 Trial
30-day money-back · cancel in one tap · how it works