4.11 Formal conditional probability and independence
Studying this course? Keep your progress in one place.
Key facts
- and .
- in general.
- .
- Independent events: , equivalently .
- Test independence with a numerical comparison, and state the conclusion. Mutually exclusive () is not the same as independent.
The formal definition
The probability of given is the fraction of that is also in .
Knowing that has happened reduces the sample space to . We then ask what fraction of is also in .
Order matters: and are usually different.
Translate words first: 'given that', 'if' and 'of those who' all introduce the conditional probability condition after the bar.
- Conditional probability
Worked example
, and . Find and .
- 1
For the condition is , so divide by : .
- 2
For the condition is , so divide by : .
and . What is ?
The multiplication form and branches
Multiply along branches for joint probabilities, add the routes to an event, then divide to reverse the condition.
Rearranging the definition gives the multiplication rule. It is how you work along the branches of a tree diagram.
If it rains with probability 0.35 and Kofi is late with probability 0.6 when it rains, .
Add the routes to find the overall probability: . Then reverse the condition with the definition: .
- 1
Multiply along branches
to get each joint probability, e.g.
- 2
Add the routes to L
to get
- 3
Divide
the joint route by to get
- Multiplication rule
and . What is ?
Working with complements
Use and .
Many questions need . Using and :
Also, , because given , either happens or it does not. The complement goes on , not on the condition.
- Overlap with B′
- Complement of B
- Complement of A given B
Worked example
, and . Find .
- 1
.
- 2
.
- 3
Divide: .
Which expression equals ?
Independent events
Events are independent when knowing one happened does not change the probability of the other.
and are independent if , equivalent to .
To test for independence, calculate one side of any of these and compare numerically. Conclude 'independent' only if they are equal, and state the numbers you compared. For example , so and are not independent.
Do not confuse this with mutually exclusive events, for which . If is given with , substitute and solve for an unknown.
Independent
Mutually exclusive
- Cannot be independent if both are non-zero
, and . Are and independent?
Conditional probability from counts
Divide by the size of the condition group, not the total.
When data are given as counts (a two-way table), divide by the size of the condition group. In a survey of 200 students where 120 travel by bus and 40 of those play tennis, . The denominator is the number in , not 200.
For independence you would need .
Interpret your conclusion in context, for example 'bus users are less likely to play tennis'.
Worked example
In the survey above, 80 students play tennis in total. Are T and B independent?
- 1
Independence needs n(T ∩ B) = 48.
- 2
The actual count is 40, and 40 ≠ 48.
- 3
So the events are not independent. Since 40 < 48, bus users are less likely to play tennis.
Of 50 students who study French, 10 also study art. What is P(art | French)?
Try an exam question
, and . (a) Find . (b) Find . (c) Determine whether and are independent.
[4 marks]
- [1].
- [1] and .
- [1].
- [1], so and are not independent.
That's the notes covered.
Carry on to the next subtopic.
Try a question
Find .
1 mark, marked against the mark scheme
Find .
1 mark, marked against the mark scheme