Chapter 6 - PPTs
Chapter 6 - PPTs
Chapter 6: Probability
➢ Bayes’s Law
Probability
It is often necessary to "guess" about the outcome of an event in order to make a
decision:
➢Politicians study polls to guess their likelihood of winning an election.
➢Doctors choose the treatments needed for various diseases based on their
assessment of likely results
➢You may have chosen your course of study based on the probable
availability of jobs
Requirements:
Given a sample space 𝑆 = {𝑂1 , 𝑂2 , … , 𝑂𝑘 }, the probabilities assigned to the outcomes must
satisfy two requirements:
➢ The probability of any outcome must lie between 0 and 1. That is,
𝟎 ≤ 𝑷 𝑶𝒊 ≤ 𝟏, 𝐟𝐨𝐫 𝐞𝐚𝐜𝐡 𝟏 ≤ 𝒊 ≤ 𝒌
➢ The sum of the probabilities of all the outcomes in a sample space must be 1. That is,
𝒌
𝑷 𝑶𝒊 = 𝟏
𝒊=𝟏
Approaches to assigning probabilities
Rule 1: Relative Frequency Approximation of Probability (experimental probability)
Conduct (or observe) a procedure a large number of times, and count the number of
times that event 𝐴 actually occurs. Then 𝑃(𝐴) is estimated as follows:
Example: Determine 𝑃(the tack lands point up), when tossing a tack.
Approaches to assigning probabilities
number of outcomes A occured
𝑃 𝐴 =
total number of times trial is repeated
Approaches to assigning probabilities
Rule 2: Classical Approach to Probability (theoretical probability)
Assume that a given experiment has 𝑛 different simple events, each of which has an
equal chance (equally likely) of occurring. If event 𝐴 can occur in 𝑠 of these 𝑛 ways,
then
number of outcomes for event A 𝑁(𝐴) 𝑠
𝑃 𝐴 = = =
otal number of possible outcomes 𝑁(𝑆) 𝑛
where 𝑁(𝐴) is the number of outcomes in event 𝐴
Example:
➢ Determine P(head), when tossing a fair coin. [P(head) = 1/2]
Example:
➢Determine 𝑃(rain tomorrow), meteorologists use their expert knowledge of
weather conditions.
➢Another example is to ask New York Yankees fans, before the baseball season
starts, about the chances of New York winning the World Series. While there is no
absolute mathematical proof behind the answer to the example, fans might still
reply in actual percentage terms, such as the Yankees having a 25% chance of
winning the World Series.
Probability of an event
• The probability of an event is the sum of the probabilities of the simple events that
constitute the event.
Example. Assume that the grades 𝐴, 𝐵, 𝐶, 𝐷 and 𝐹 on a statistics test have been recorded
and the relative frequencies are computed as follows:
𝑃(𝐴) = 0.20, 𝑃(𝐵) = 0.30, 𝑃(𝐶) = 0.25, 𝑃(𝐷) = 0.15, 𝑃(𝐹) = 0.10
Find the probability of the event, pass the course.
Note. No matter what method was used to assign probability, we interpret it using the relative
frequency approach for an infinite number of experiments.
Example
Assume that two dice have been rolled.
1. List all the elements of the sample space
2. Find the probability that the sum of two outcomes is 5.
Joint, marginal and conditional probabilities
➢ Intersection of Events A and B (joint probability)
• The intersection of events 𝐴 and 𝐵 is the event that occurs when both
𝑨 𝐚𝐧𝐝 𝑩 occur. It is denoted as
𝐴 and 𝐵, (for this situation it means; 𝑨 𝐣𝐨𝐢𝐧𝐭 𝑩)
𝑃 𝐴′ = 1 − 𝑃(𝐴)
Compound Event: A compound event is any event combining two or more simple events.
Some formulas
➢ Complement rule:
𝑃 𝐴′ = 1 − 𝑃(𝐴)
Example
➢ Let’s flip a coin. Find the probability that the outcome is either head or tail.
➢ Now, let’s flip a coin twice (or flipping 2 coins at the same time). What is the probability
of getting at least one head?
Example
Consider a standard deck of 52 cards. We select a card randomly from the deck. Find the
probability that the selected card is either diamond or Queen.
Example
The Humber Bookstore has found that in the past 35% of sales include notepads, 75% include
textbooks and 20% include both textbooks and notepads.
1. What is the probability that a sale will include notepads or textbooks?
2. Determine the probability that a sale will include textbooks but not notepads.
3. Determine the probability that a sale will include not notepads and not textbooks.
Example
There are 30 female students and 20 male students in a STAT class.
1. What is the probability to select one student who is a female student?
2. What is the probability to select one student who is not a female student?
Example
There are 5 red gumballs, 4 blue gumballs, 3 yellow gumballs and 2 green gumballs in a
gumball machine.
1. What is the probability of choosing a red gumball?
2. What is the probability that a red gumball will not be chosen?
Contingency table (joint probabilities)
Contingency tables are a great way to classify outcomes and calculate different types of
probabilities. These tables contain rows and columns that display bivariate frequencies of
categorical data.
In a survey, the following data are collected from experimental and control groups to test
Advil.
joint marginal
probabilities probabilities
➢ Dependent outcomes: If 𝐴 and 𝐵 are not independent, they are said to be dependent.
a) Randomly select ONE person. What is the probability that the selected person
has headache and uses Advil.
Solution.
Multiplication rule
Consider the following contingency table:
Advil Placebo Control Group Totals
Headache 50 48 24 122
No Headache 732 616 602 1950
Totals 782 664 626 2072
b) Randomly select TWO persons with replacement. What is the probability that
the selected persons are from control group.
Solution.
Multiplication rule
Consider the following contingency table:
Advil Placebo Control Group Totals
Headache 50 48 24 122
No Headache 732 616 602 1950
Totals 782 664 626 2072
Solution.
Conditional Probability
➢ A conditional probability of an event applies when the probability is affected by the
other events.
➢ 𝑃(𝐵|𝐴) represents the probability of 𝐵 occurring after it is assumed that the event 𝐴
has already occurred.
➢ 𝑃(𝐵|𝐴) (We can read 𝐵|𝐴 as “𝐵 given 𝐴”) can be found by dividing the probability of
events 𝐴 𝐚𝐧𝐝 𝐵 both occurring by the probability of event 𝐴.
Keywords:
Given that …, Suppose that …, Assume that …
Example
Consider the following contingency table:
Advil Placebo Control Group Totals
Headache 50 48 24 122
No Headache 732 616 602 1950
Totals 782 664 626 2072
a) If a person is randomly selected, what is the likelihood that he/she got headache,
given that the person in placebo group is selected?
b) Given that a person in headache group is selected, what is the probability that
the person got Advil?
Example
Solution a) Advil Placebo Control Group Totals
Headache 50 𝟒𝟖 24 122
No Headache 732 616 602 1950
Totals 782 𝟔𝟔𝟒 626 𝟐𝟎𝟕𝟐
Example
Solution b) Advil Placebo Control Group Totals
Headache 𝟓𝟎 48 24 𝟏𝟐𝟐
No Headache 732 616 602 1950
Totals 782 664 626 𝟐𝟎𝟕𝟐
Example
Consider the following contingency table:
Advil Placebo Control Group Totals
Headache 50 48 24 122
No Headache 732 616 602 1950
Totals 782 664 626 2072
If a person is randomly selected, what is the likelihood that he/she got no headache, or
Advil?
If a person is randomly selected, what is the likelihood that he/she is in Placebo or Control
groups?
Solution. Two events are mutually exclusive:
Example – Practice at home
Suppose a study of speeding violations and drivers who use cell phones produced the following fictional
data:
• These diagrams are helpful in counting the number of possible outcomes if the
number of possible is not too large.
P(1 Boy and 1 Girl) = P(B and G) + P(G and B) = 0.25 + 0.25 = 0.50
Example
An insurance company survey shows that 90% of Ontario drivers/passengers wear seatbelt. For
those who wear seatbelt, it is estimated that the chance of getting a serious injury in a car
accident is 20% whereas those who do not have a 60 % chance of getting a serious injury.
a) Draw a probability tree diagram. Show all possible outcomes and probabilities.
b) What is the probability that the driver/passenger suffers serious injury in a car
accident?
c) If there is a serious injury, find the probability that the driver/passenger does not wear
seatbelt.
b)
𝑃 serious injury = 𝑃 seatbelt and serious injury + 𝑃 no seatbelt and serious injury
= 0.18 + 0.06 = 0.24
c)
𝑃 no seatbelt joint serious injury 0.06
𝑃(no seatbelt | serious injury) = =
𝑃 serious injury 0.06 + 0.18
0.06
= = 0.25
0.24
Bayes’s law (Optional)
Bayes’s theorem (Bayes's law or Bayes's rule) describes the probability of an event, based on prior
knowledge of conditional probabilities that might be related to the event.
Reading Materials.
Please study Chapter 6 – Probability from the official textbook.
Also, please see extra examples at the end of Chapter 6 (notes) posted on the
course page.
You may wish to practice the following suggested problems from Chapter 6