1.8. Concepts Recap#
In this section we practice using formal notation for probability.
Example:
what is meant by the term ‘mutually exclusive’? Include formal notation and a real life example in your answer.
Answer:
Two events A and B are mutually exclusive if they cannot both occur together in a sample
Put formally, p(A|B) = 0, p(B|A) = 0, and p(A∩B) = 0
For example if I stop a random person on the street:
Let A be the event that that person has blue eyes
Let B be the event that that person has brown eyes
A and B are mutually exclusive since a person cannot have both blue eyes and brown eyes.
1.8.1. Definitions#
Define the following terms, in each case including any relevant formal notation, and give a real life example:
a) statistically independent events
Click to reveal answer
Two events A and B are statistically independent if the probability of A does not depend on whether B occurred, or vice versa.
Put formally, \(p(A|B) = p(A)\) and \(p(B|A) = p(B)\)
For example, if I stop a random person on the street
Let A be the event that the person has blue eyes
Let B be the event that the person is male
A and B are statistically independent as eye colour does not depend on sex.
Note – it is not really correct to phrase this as “the probability of A does not depend on the probability of B”. Rather, the probability of A does not depend on whether B is true in this particular instance which is not a probability, it’s an event.
b) \(A\) and its complement \(A^c\)
Click to reveal answer
The complement of A is the event that A does not occur.
\(A\) and \(A^C\) are mutually exclusive
For example if I pick a random student, let \(A\) be the event that she is studying psychology. Then \(A^C\) is the event that she is NOT studying psychology.
c) What is the multiplication law?
Use formal notation and the appropriate symbol \(\cup\) or \(\cap\)
Click to reveal answer
In words: the probability that events \(A\) and \(B\) both occur is the probability that \(A\) occurs, times the probability that \(B\) occurs given \(A\) occurs.
d) If two events are statistically independent, how can the multiplication law be simplified?
Click to reveal answer
If \(A\) and \(B\) are statistically independent then \(p(A|B) = p(A)\) and \(p(B|A) = p(B)\)
Then the multiplication law becomes
e) What is the addition law?
Use formal notation and the appropriate symbol \(\cup\) or \(\cap\)
Click to reveal answer
f) If two events are mutually exclusive, how can the addition law be simplified?
Click to reveal answer
If the events \(A\) and \(B\) are mutually exclusive then \(p(A \cap B)=0\) so the addition law becomes:
1.8.2. Combinations of events#
In 2002, a social survey with 1117 respondents in total found that 96 were members of an environmental group. Of these, 30 said they would be willing to pay higher prices for their shopping, to protect the environment. Of the remaining respondents, 88 said they would be willing to pay higher prices to protect the environment.
a) Construct a contingency table for these data
Click to reveal answer
{image} https://raw.githubusercontent.com/jillxoreilly/StatsCourseBook_2024/main/images/Chp9_ExerciseTable.png :width: 80% :align: center
Let \(E\) be the event that the respondent is a member of an environmental group
Let \(P\) be the event that the respondent is willing to pay higher prices to protect the environment
Answer using formal notation as in this example
Example: Are \(E\) and \(P\) mutually exclusive?
Answer: No, if \(E\) and \(P\) were mutually exclusive, \(p(E \cap P)\) would be zero, but in fact \(p(E \cap P) = \frac{30}{1117}\).
a) Are \(E\) and \(P\) statistically independent?
Click to reveal answer
No, if they were independent then \(p(P|E)\) should be equal to \(p(P)\).
In fact \(p(P|E) = \frac{30}{96} = 0.31\) and \(p(P) = \)frac{118}{1117} = 0.11$
b) What is the probability that a member of an environmental group would be willing to pay higher prices?
Click to reveal answer
This is \(p(P|E) = \frac{30}{96} = 0.31\)
c) What is the probability that a non-member would be willing to pay higher prices?
Click to reveal answer
This is \(p(P|E^c) = \frac{88}{1021}\)
1.8.3. Bayes Theorem#
A rapid screening test for COVID19 has the following efficacy statistics:
For people with COVID19, a positive result occurs 90% of the time
For people without COVID19, a positive test result occurs 2% of the time
In the city of Easton, it is estimated that 10.7% of the population have COVID.
a) A person in Easton takes a COVID test and it comes out positive. What is the chance this person has COVID?
Click to reveal answer
Let \(T\) be the event of a positive covid test
Let \(C\) be the event that a person actually has COVID
We need p(C|T). From Bayes Theorem:
We know \(p(T|C) = 0.9\) We also know \(p(C) = 0.107\) (base rate of COVID infection in Easton)
We can work out \(p(T)\) by considering the two cases:
plugging in numbers:
Now we return to Bayes’ Theorem:
The chance the person has COVID is 84%
In the city of Westerby, it is estimated that 0.5% of the population have COVID.
b) A person in Westerby takes a COVID test and it comes out positive. What is the chance this person has COVID?
Click to reveal answer
By the same method as above:
We know \(p(T|C) = 0.9\) We also know \(p(C) = 0.005\) (base rate of COVID infection in Westerby)
And using to Bayes’ Theorem:
The chance the person has COVID is 18%
c) Comment on the results of parts a) and b)
Click to reveal answer
The same positive COVID test has a very different interpretation in an environment where COVID is prevalent (the base rate or prior probability is high, around 10%), compared to an environment where COVID is rare.