BUSINFO 705 Chap.3 Discrete and Continuous Probability Models
Discrete and Continuous Probability Models
This chapter chooses probability models from the data-generating process. It compares discrete uniform, binomial and Poisson counts with continuous uniform and normal measurements. The worked binomial calculation makes every assumption and parameter explicit, while later examples scale a Poisson mean to the requested interval and translate a normal value into a z-score.
A final clinic targets support, interval and tail-direction errors before they reach a decision. Model choice begins with the variable, stopping rule and support rather than a familiar business noun. Sketches and directional checks provide a second line of defence against function and complement errors.
What this chapter covers
- 01
Discrete uniform outcomes
- 02
Binomial trials and combinations
- 03
Poisson counts and interval scaling
- 04
Continuous uniform interval probability
- 05
Normal location and spread
- 06
Z-scores and tail direction
Worked example · free
Find an exact binomial probability
- 1Define success as acceptance and identify n=4, x=2 and p=0.70.
- 1Count the arrangements: C(4,2)=6.
- 2Substitute P(X=2)=6(0.70)²(0.30)².
- 2The probability is 0.2646, or 26.46%, subject to fixed trials, constant p and independence.
Key terms
- Discrete uniform
- A distribution assigning equal probability to each value in a finite discrete support.
- Binomial distribution
- A model for success counts across fixed independent trials with constant success probability.
- Poisson distribution
- A count model for arrivals during a stated span when average intensity remains stable.
- Continuous uniform
- A bounded continuous distribution with equal density across its support.
- Normal distribution
- A symmetric bell-shaped continuous distribution defined by its mean and standard deviation.
- Z-score
- The number of standard deviations a value lies above or below a mean.
Discrete and Continuous Probability Models FAQ
How do binomial and Poisson models differ?
Binomial models successes among a fixed number of opportunities and needs n and p. Poisson models event counts over a fixed interval and needs the interval mean. Both produce counts, but their process assumptions and supports are different.
Why must a Poisson mean be rescaled?
The parameter belongs to a stated interval. If the mean is six events per hour, a stable process has mean three per half-hour. Using the hourly value in a shorter-interval question models a different event count and distorts the probability.
How can I avoid a normal-tail mistake?
Sketch the bell curve, mark the boundary relative to the mean and shade the requested side before using a function. A cumulative function typically returns the left tail, so a right-tail event requires subtracting that value from one.
When is a continuous uniform model appropriate?
It fits a quantity bounded by known endpoints when equal density across that range is defensible. Compute an interval probability as its length divided by the total support length. If observations cluster near a centre or some ranges are more plausible, the flat-density assumption needs reconsideration.
How do I check a normal model against physical limits?
Compare the curve's implied range with the process. A normal model assigns some probability to every real value, including negative values that may be impossible. Assess whether that impossible tail is negligible; if it is material, use or test a bounded alternative rather than ignoring the conflict.
Assessment move
Build a one-page comparison table with support, process, parameters and one diagnostic question for each model. Practise recognising a distribution from words before touching a formula. For binomial problems, list n, x and p and explain why the trials are fixed and independent. For Poisson problems, write the interval beside the mean and rescale it first.
For continuous uniform questions, calculate area as an interval-length ratio. For normal questions, sketch and shade the event, compute the z boundary and check whether the returned tail direction is plausible. Keep the Quiz 1 exclusion of exponential and triangular models separate from the wider lecture content.
Take one count context and rewrite its stopping rule so it changes from fixed trials to a time interval; explain why the model changes. For every spreadsheet function, write the event in symbols and predict whether the answer should be below or above one half. Create a model-choice card for each distribution with four fields: variable, support, mechanism and parameters.
Include one counterexample showing when that model would fail. Practise scaling a Poisson mean from hours to minutes and days, keeping the event interval beside the parameter. For a continuous uniform model, clip a requested interval to the support before taking the length ratio. For a normal model, locate the boundary relative to the mean, shade the event and test the returned tail direction.
Calculate whether an impossible negative region is negligible when using a normal approximation to a positive quantity. Finally, compare exact-point probability for discrete and continuous variables and explain why the latter is zero. These checks expose classification, interval and complement errors before arithmetic dominates the answer.
Working through Discrete and Continuous Probability Models in BUSINFO 705? Sia is AskSia’s AI Management tutor — ask any BUSINFO 705 Discrete and Continuous Probability Models question and get a clear, step-by-step explanation grounded in how BUSINFO 705 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.