ST 371 Note Outline 2: Probability
Textbook: 2.1, 2.2, 2.4 and 2.5
Sample Space, Events, and Probability
_______________________________: Any action or process whose outcome is
subject to uncertainty
Ex:
Ex:
Ex:
_______________________________: The set of all possible outcomes of an
experiment
Notation:
Ex: (Flipping a coin)
(Rolling a dice)
Ex: (Taking 3 free throws)
Note:
Ex: (Breaking Strength of Cotton Thread)
_______________________________: Any collection of outcomes from the sample
space
Notation:
Page 1 of 13
Ex: (Taking 3 free throws)
_______________________________ of event A: Set of all outcomes in S that are
not in A
Notation:
Ex: (Taking 3 free throws)
_______________________________ of two events A and B: Event consisting of all
outcomes that are in both A and B
Notation:
Ex: (Rolling a dice) Let A = {1, 2, 3} and B = {1, 3, 5}
Ex: (Taking 3 free throws) Let A and B be defined as on the previous
page
_______________________________ of two events A and B: Event consisting of all
outcomes that are either in A or in B or in both (i.e. in at least one event)
Notation:
Ex: (Rolling a dice) Let A = {1, 2, 3} and B = {1, 3, 5}
Page 2 of 13
Ex: (Taking 3 free throws) Let A and B be defined as on the previous
page
______________________________________________: A and B are said to be
mutually exclusive (or disjoint) if their intersection is empty
_________________________: A precise measure of the chance that a particular
event will occur
Notation:
Note: One way to think about probability is as a long-run average or
rate: Probability represents the
Ex: Consider an experiment where we flip a single, fair coin one time.
Recall that for such an experiment, S = {H, T} (where H=Heads and
T=Tails). What is the probability of getting a headsi.e. what is P(H)?
Suppose now that I actually carry out the experiment. After one flip,
what percentage of the time did I get heads?
What other possibility was there?
What possibilities are there after two flips?
What possibilities are there after three flips?
Note:
Page 3 of 13
The graph below illustrates this long-run nature of probability:
Percent
heads
Number of coin
flips
Ex: The local Meteorological claims that there is a 70% probability of
rain tomorrow. Select the best interpretation of this statement:
a. Approximately 70% of the city will receive rain within the next 24
hours.
b. Historical records show that it has rained on 70% of previous
occasions with the same weather conditions.
c. If we were to repeatedly monitor the weather tomorrow, 70% of
the time it will be raining.
d. Historical records show that it has rained on 7 of the 10 previous
occasions with the same weather conditions.
The interpretation of probability as a long-run rate (rather than a
prediction) is important to keep in mind when considering actual
numeric values of probability:
Events with a high probability (_______________________________) are
not guaranteed to occur, but they are
_______________________________ to occur.
Events with a low probability (_______________________________) could
still happen, but they are _______________________________ to occur.
Page 4 of 13
Probabilities based on _________________________________________ are
not as reliable as those based on
_________________________________________.
Page 5 of 13
Core Properties of Probability (Axioms):
1. For any event A,
2.
3. If A1, A2, is an infinite collection of disjoint events, then
Note: Together, these imply that property (3) holds for a
_______________________________ collection of disjoint events (see the
Proposition on page 56 of the textbook).
Additional Properties of Probability (Propositions):
1. For any event A,
2. For any event A,
3. For any two events A and B,
Note: This can be extended to three or more events as follows:
Conditional Probability
Page 6 of 13
Consider two events A and B. What is the probability that A will occur given
that we know B has occurred? Or, what is the probability of A conditioned on
B?
Ex: Consider an experiment where three names are drawn from a hat.
Let A = your name is first and B = your name is one of the three.
Ex: Now suppose the experiment is a drug test for a randomly
selected individual. Let A = person used the drug and B = person
tests positive for the drug.
___________________________________________________: For any two events A and
B with P(B) > 0, the conditional probability of A given that B has occurred is
Notation:
Note: The key idea of a conditional probability is that
Page 7 of 13
EX: (Drug Test)
Test +
0.02
0.01
Drug
No
Drug
Test 0.01
0.96
___________________________________________________: Solving the condition
P( A B)
probability equation for
gives
Ex: Suppose that when you play a video game for the first time, you
have a 0.3 probability of winning. Your probability of winning is higher
when you play for a second time, but how much higher depends on
your first result. If you win on the first game, then your probability of
winning on the second game increases to 0.8. But if you lose on the
first game, then your probability of winning on the second game
increases only to 0.4.
What is the probability that you win on both of your first two plays?
What is the probability that you win on your second play of the game, regardless of
whether or not you won on your first play?
Page 8 of 13
The previous example is an illustration of the
________________________________________________________________: Let A1, ,
Ak be mutually exclusive and exhaustive events. Then for any other event B,
P ( B ) P( B | A1 ) P( A1 ) L P ( B | Ak ) P ( Ak )
k
P ( B | Ai ) P ( Ai )
i 1
This law allows us to calculate probabilities of events by conditioning on
other events. Sometimes these conditional probabilities are easier to
calculate, i.e. it may be easier to find P(B|A) than P(B). Similarly, there are
times when it may be easier to calculate P(B|A) than P(A|B). Fortunately,
there is another law that can be used to help in this situation.
___________________________________________________: Let A1, , Ak be a
collection of k mutually exclusive and exhaustive events with prior
probabilities P(Ai) (i=1,k). Then for any other event B for which the P(B) >
0, the posterior probability of Aj given that B has occurred is
P( Aj | B)
P( Aj B )
P( B)
P( B | Aj ) P( Aj )
k
P( B | A ) P( A )
i 1
j=1,...,k
Ex: (Video Games) Given that you won on your second play of the game, what is the
probability that you won on your first play?
Page 9 of 13
Given that you lost on your second play of the game, what is the probability that you won
on your first play?
Ex: (Drug test) Drug tests are designed with a known sensitivity. However, those
interested in such tests want to know the probability that the person tested actually used
the drug given that the test was positive. Using appropriate notation, write out and solve
for the probability of interest. (Assume the sensitivity and specificity are both 99% and
0.5% of the population uses the drug.)
Page 10 of 13
Independence
Conditional probability tells us about the relationships between events. If
P(A|B) is very different than P(A), then knowing B tells us a lot about A. But
what if P(A|B) is not very different than P(A)? Then knowing B does not give
much information about A.
Two events A and B, are _________________________ if:
Otherwise the events are _________________________.
Logically: Let A = tomorrows high temperature in Raleigh and B =
todays high temperature in Paris.
A and B are _________________________ since
Let C = tomorrows high temperature in Durham. A and C are
_________________________ since
Proposition: A and B are independent if and only if:
Ex: (Video Game) We could represent the probabilities from the video
game example in the table below. Use this to determine if the
outcomes of the two turns are independent.
Win 2nd Turn
Lose 2nd
Turn
Total
Win 1st Turn
0.24
0.06
0.3
Lose 1nd Turn
0.28
0.42
0.7
Total
0.52
0.48
1.00
Page 11 of 13
But what if these were the probabilities:
Win 2nd Turn
Lose 2nd
Turn
Total
Win 1st Turn
Lose 1nd Turn
Total
This proposition extents to more than two events: Events A1, , An are
mutually independent if, for every k (k=2,3,n) and every subset of indices
i1,i2,ik,
P( Ai1 Ai2 L Aik ) P( Ai1 ) P( Ai2 ) L
P ( Aik )
Ex: A multiple choice test has 20 questions, each with four possibilities.
If a student has an 80% chance of getting each question correct, and
the questions are independent, what is the probability that student will
get a perfect score?
Let Aj = students answer is correct on question j.
Page 12 of 13
Page 13 of 13
