ECON 225: Data and

Statistics for Economics

Lecture 7
Probability – outline of topics
1. The notion of randomness and probability
• Concept of experiments, independent trials
2. Probability model
• Sample space, event space, and probability function
3. Multiplication rule for independent events
4. Addition rule
5. Conditional probability and Bayes’ rule
Plan for today
• The probability model
• Three axioms of probability
• The complement rule
• Excel: boxplots (Q 1.50 in Alwan)
Why study randomness and probability
• A phenomenon is random if individual outcomes are uncertain but
there is a regular distribution of outcomes in a large number of
repetitions.
• When we draw a sample from a population, the observations in the
sample are random
Concept of probability
• Random experiment: Process that leads to one of several possible
outcomes
• The probability of any outcome of a random experiment can be
defined as the proportion of times the outcome would occur in
infinitely many repetitions of the experiment
Coin toss

What are the

outcomes of this
100%
experiment?
head or tail
Coin toss
• The result of any single coin toss experiment is random. But the result
over many tosses is predictable as long as the experiments, or trials,
are independent (i.e. the outcome of a new coin flip is not influenced
by the outcome of the previous flip.)
• On the previous slide, we see that as the number of flips increases,
the proportion of heads among all the flips approaches 0.5.
• We say that the probability of heads is 0.5.
Independence
• Two experiments/trials are independent if the outcome of one
trial does not influence or change the outcome of the other.
• When are trials not independent?
• Imagine that a pile of coins were spread out so that half were
heads up and half were tails up. Close your eyes and pick one. The
probability of its being heads is 0.5. However, if you don’t put it
back in the pile, the probability of picking up another coin and
having it be heads is now less than 0.5.
• The trials are independent only when you put the coin back each
time. This is called sampling with replacement.
Examples of independent trials
• Flipping coins
• Tossing dice
• Drawing cards from a shuffled deck (replacing the card each time)
• Later, we will see a mathematical definition of independence that will
help us determine if trials are independent or not
Probability models
• To formalize ideas, we will talk about probability models
• A probability model has the following parts:
• Sample space
• Event space
• Probability function
Sample space
• The sample space (S) is a set, or list, of all possible outcomes of a
random experiment.
• 𝑆 = {𝑂!, 𝑂# , … , 𝑂$ }
• Note that the sample space is
• Exhaustive: lists all possible outcomes
• Mutually exclusive: the experiment can only produce one of the
outcomes in the list
• When rolling a die, the sample space is: S = {1,2,3,4,5,6}
 Event
• Event: a subset of the sample space
• In other words, an event is some combination of outcomes
• Example: the event of rolling a 5 or a 6
• Event space: the set of all subsets of the sample space; the
set of all possible events
Space : {1, 2, 3, 4, 5, 6}
Note:
or = +
and = x
Probability function
• A probability function assigns a number between 0 and 1 to
each event (i.e. to every element of the event space)
• This number is called the probability of the event
• If an event has probability zero, it definitely will not happen
• If an event has probability strictly between zero and one, it
might happen
• If an event has probability one, it definitely will happen
,

Probability model for a coin toss

• Sample space: S = { H, T }
• Event space: F = { {}, {H}, {T}, {H,T}}
• Probability function:
• P({}) = 0 NOTE: Formula for sample space: 2^n

• P({H}) = 0.5 P = 1/2

• P({T}) = 0.5 P = 1/2
• P({H,T}) = 1 P = 1/2 + 1/2 = 1
OR
Example 1
Consider a probability model consisting of randomly drawing 2 colored
balls (without replacement) from a jar containing 2 red balls and 1 blue
ball. If we assume B = blue and R = red, the sample space of this
experiment is:
a) S = {B, R}.
b) S = {BR, RB}.
c) S = {BB, BR, RB, RR}.
d) S = {RR, BR, RB}.
 It is the question that determines the sample
space
A basketball player shoots three
free throws. What are the
possible sequences of hits (H)
and misses (M)?

A basketball player shoots three

free throws. What is the number S = {0,1,2,3}
of baskets made?
Axioms of probability 1
A function defined on the event space must follow 3 rules for
it to be a probability function
1. Probabilities range from 0 (no chance of the event) to 1 (the event is
certain to happen)
• For any event A, 0 ≤ P(A) ≤ 1
• Example: Probability of getting a Head = 0.5
• We write this as: P(Head) = 0.5
• P(neither Head nor Tail) = 0
Axioms of probability 2
Definition: Two events A and B are called
disjoint if if they have no outcomes in common
and can never happen together

2. If A and B are disjoint events, the probability

that A or B occurs is the sum of their
individual probabilities. This is called the
addition rule for disjoint events.
Example 2
If you flip two coins, what is the sample space? What is the
probability that you obtain “only heads or only tails” ?
S = {HH, HT, TH, TT}
P {(HH, TT)} = 1/4 + 1/4 = 0.5
Axioms of probability 3
3. Because some outcome must occur on every trial, the sum
of the probabilities for all possible outcomes of the sample
space must be exactly 1.
• P(S) = P(sample space) = 1
• Example: S = {Head, Tail}
• P(head) + P(tail) = 0.5 + 0.5 =1 à P({head, tail}) = 1
Complement rule
• Definition: The complement of any event A is
the event that A does not occur, written as
Ac.
• From the axioms of probability, we can
deduce the complement rule:
• The probability of an event not occurring is 1
minus the probability that it does occur
• Mathematically: P(not A) = P(Ac) = 1 − P(A)
Complement rule
• Everything in the sample space is either in A or not in A.
• P(S) = P(A or Ac)
• A and Ac are disjoint events. By Axiom 2,
• P(A or Ac) = P(A) + P(Ac)
• By Axiom 3, P(S) = 1
• Therefore, 1 = P(A) + P(Ac)
• Rearranging, we get the complement rule: P(Ac) = 1 - P(A)
Assigning probabilities: Finite number of
outcomes
Finite sample spaces deal with discrete data—data that can
take on only a limited number of values. These values are
often integers or whole numbers.

The individual outcomes of a random phenomenon are

always disjoint. è The probability of any event is the sum of
the probabilities of the outcomes making up the event
(addition rule).
Probabilities: Equally likely outcomes
We can assign probabilities either:
• empirically à from our knowledge of numerous similar past events
• Mendel discovered the probabilities of inheritance of a given trait
from experiments on peas without knowing about genes or DNA.
• Or theoretically à from our understanding of the phenomenon and
symmetries in the problem
• A six-sided fair die: each side has the same chance of turning up
• If a random experiment has k equally likely outcomes, each individual
outcome has probability 1/k
Example 3
You toss two die. What is the probability of the outcomes summing to
5?
outcome = 5
{(1,4), (2,3), (3,2), (4,1)}

P(5) = 4/36 = 1/9

Example 3
You toss two die. What is the probability of the outcomes summing to
5?
S can be written:
{(1,1), (1,2), (1,3),…,
(5,6), (6,6)}

There are 36 possible

outcomes, all equally
likely. The probability of
each one is 1/36.

P(sum is 5) = P(1,4) + P(2,3) + P(3,2) + P(4,1) = 1/36 = 1/9

4/36
For next class
• Reading: Alwan 4.3
• For practice: 4.67, 4.93, 4.95
• Requirements for group project have been posted to Canvas