Modelling with probability
Studying this course? Keep your progress in one place.
Key facts
- A model is a simplified description of a situation built from assumptions; answers are only as good as the assumptions.
- Common assumptions: a fair die or coin, independent trials, constant probability.
- Criticise an assumption by naming it, saying why it is doubtful here, and stating the likely effect.
- A binomial model needs fixed , two outcomes, constant and independent trials.
- Compare expected with observed results to test a model.
What a probability model is
A model uses assumptions to let you calculate probabilities.
A model is a simplified description of a real situation, built from assumptions. Common ones are that a coin or die is fair, that trials are independent, and that the probability stays constant.
Modelling two die rolls with a fair die gives . All the working in a model question is done inside the model.
- 1
Real situation
Describe what is happening.
- 2
Assumptions
Fair, independent, constant probability.
- 3
Model
Choose a distribution or sample space.
- 4
Calculate
Work out the probabilities.
- 5
Check
Compare with the data and criticise.
Which is an assumption of the fair-die model?
Critiquing assumptions
Name the assumption, say why it is doubtful in this context, and state the likely effect.
Be specific to the context. Real dice and coins can be biased. Independence fails when a common cause links events, such as traffic delaying consecutive buses. Constant probability can vary with time, weather or form.
A good answer names the assumption, says why it is doubtful here, and states the likely effect.
Assumption
- Fair die or coin
- Independence
- Constant probability
Why it may fail
- Real ones can be biased
- A common cause links events
- It varies with time, weather or form
A model assumes a bus is late with the same probability each day. Which criticism is best?
Using distributions to model situations
Check the conditions before choosing: a binomial model needs a fixed number of independent trials with constant .
A binomial model needs a fixed number of trials, two outcomes, constant and independence. A discrete uniform model suits a fair die, giving each value .
A discrete model given as a table is valid if its probabilities sum to 1. For a continuous variable, probabilities are areas under a curve.
- Binomial
Worked example
. Find and .
- 1
.
- 2
.
- 3
Add the three probabilities.
Which situation is a binomial model?
Comparing a model with data
Compute the expected result and compare it with what was observed.
The expected value is the result predicted by the model, in trials. The observed value is what actually happened.
A fair die rolled 60 times gives an expected 10 sixes; 18 sixes is far more, so the ‘fair’ assumption looks doubtful. Small samples give weak evidence, so say a larger sample would help.
- Expected number
A fair die is rolled 60 times. How many sixes do you expect?
Try an exam question
The number of days out of on which a bus is late is modelled by . State two assumptions of this model and find .
[4 marks]
- [1]Each day is independent of the others.
- [1]The probability of being late is constant at .
- [1].
- [1] (3 s.f.).
That's the notes covered.
Carry on to the next subtopic.
Try a question
Using the model, find the probability that the total of the two rolls is 7.
1 mark, marked against the mark scheme
Using the model, find the probability that the total of the two rolls is 7.
1 mark, marked against the mark scheme