1 / 39 2 / 39
Notation
Usually capital letters for sets: A, B, C.
Union: x ∈ A ∪ B ⇐⇒ x ∈ A or x ∈ B.
Lower case for elements of sets: x, y, z.
x ∈ A means “x is an element of A” Intersection: x ∈ A ∩ B ⇐⇒ x ∈ A and x ∈ B.
x∈/ A means “x is not an element of A” Universal set: set of all integers, all galaxies, all bulldogs, etc.
B = {x 2 | x = 1, 2, 3} = {1, 4, 9} Complement: x ∈ Ac ⇐⇒ x ∈
/ A.
3 / 39 4 / 39
Definitions Partition
Ai ∩ Aj = ∅ for i 6= j; Example
For i = 0, 1, 2, 3, 4, Bi = {outcomes with i heads}.
5 / 39 6 / 39
Experiments
Procedure + Observations
Real Experiments are too complicated.
Instead, we analyze/develop models of experiments.
Section 1.2: Applying Set Theory to Probability
Example An experiment consists of the following procedure,
observation and model:
1. Procedure: Flip a coin and let it land on a table.
2. Observation: Observe which side (head or tail) faces you after
the coin lands.
3. Model: Heads and tails are equally likely. The result of each
flip is unrelated to the results of previous flips.
7 / 39 8 / 39
Outcomes Example
9 / 39 10 / 39
Theorem
11 / 39 12 / 39
Axioms Consequences
Theorem 1.3:
If A = A1 ∪ A2 ∪ · · · ∪ Am and Ai ∩ Aj = ∅ for i 6= j, then
A probability measure P[·] is a function that maps events in the
m
sample space to real numbers such that X
P[A] = P[Ai ].
Axiom 1 For any event A, P[A] ≥ 0. i=1
13 / 39 14 / 39
If G is the grade and N is the score, then Given: P[V ] = 0.7 P[L] = 0.6 P[VL] = 0.35
B L
1 1 2
P[G = A] = P[N ≥ 9] = P[N = 9] + P[N = 10] = + = (a) Fill in the following probability table: V
9 9 9
D
15 / 39 16 / 39
Example 2 / Slide 2
(b) Compute the following:
17 / 39 18 / 39
19 / 39 20 / 39
Law of Total Probability
21 / 39 22 / 39
Two coins: one biased, one fair, but you don’t know which is
P[A B] P[A | B] P[B] which.
P[B | A] = =
P[A] P[A]
Coin 1: P[H] = 3/4 Coin 2: P[H] = 1/2
If B1 , B2 , . . . , Bn , is a partition, then Pick a coin at random and flip it.
Let Ci denote the event that coin i is picked.
P[A | Bi ] P[Bi ]
P[Bi | A] = Pn .
i=1 P[A | Bi ] P[Bi ] What is P[C1 | H], the probability that the biased coin was flipped,
given that the coin flip was heads?
23 / 39 24 / 39
Solution: Tree Diagram Which Coin was Flipped?
3/4 H
P[C1 H]
C1 P[C1 | H] = P[H | C1 ] = 3/4, P[H | C2 ] = 1/2 =⇒ P[C1 | H] = 3/5
P[H]
1/2
1/4
T Now compute: P[C2 | T ].
P[C1 H]
= If the flip is heads, what is your guess for which coin was flipped?
P[C1 H] + P[C2 H]
1/2 H If the flip is tails, what is your guess?
1/2 3/8 3
C2 = = Finally, compute the probability that you guess the correct coin.
3/8 + 1/4 5
1/2
T
25 / 39 26 / 39
1 3 3
3/4
H P[C1 H] = 2 · 4 = 8
27 / 39 28 / 39
Bayes’ Theorem: Test for a Rare Disease Test for a Rare Disease: Tree Diagram
29 / 39 30 / 39
31 / 39 32 / 39
Drug Testing: Tree Diagram Drug Testing: Solution
+ P[D, +]
0.99 P[D | +] =
P[+]
D (0.005) · (0.99)
0.005 =
0.01
− (0.005) · (0.99) + (0.995) · (0.01)
≈ 0.332
+
0.995 0.01 P[Correct] = P[D, +] + P[N, −]
N = (0.005) · (0.99) + (0.995) · (0.99)
0.99
− = 0.99
33 / 39 34 / 39
Independence / Slide 1
35 / 39 36 / 39
Independence / Slide 2 Example
37 / 39 38 / 39
Solution
39 / 39