Chapter 6: The Mathematics of Probability Flashcards

1
Q

❮ i,j ❯

A

Ordered Pair (Probability Calculus)

i = outcome of first roll
j = outcome of second roll.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Example: Ordered pairs ❮ 1,4 ❯ and ❮ 4,1 ❯ are __________ different ___________.

A

two; outcomes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

“S”

A

Represents all possible outcomes in a random experiment.
“Sample Space”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

“Sample Space” must be either _____________ or ____________ _____________.

A

finite; countably infinite

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Event A

A

Subset of the Sample Space (S).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

An Event can be ___ and ___, ____ or ___, or _____-___ and _____-___.

A

A and B; A or B; not-A and not-B

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Aᶜ

A

Complement of Event A

The complement of Event A is all other Events in the Sample Space.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Sᶜ

A

Complement of Sample Space (S)

Sᶜ = ∅ (where “∅” is a null sign, meaning “empty set”).

The complement of a Sample Space (S) is an empty set.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

p

A

p = Probability Measure

Takes an Event as it’s argument, and assigns a real number 0-1 as the probability of that event.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Probability Calculus (Set of 3 Axioms)

A

1) Probability of an Event is a real number between 0 and 1.

2) Probability of Sample Space (S) is 1.

3) If two Events are mutually exclusive (meaning both can’t happen simoultaneously), then the probability of one Event happening equals the probability of the first plus that of the second.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Mutually exclusive

A

both events can’t happen at the same time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Probability Calculus (Formalized)

A

1) p(A) = 1≥p(A)≥0
2) p(S) = 1
3) If A ᑎ B = ∅, then p(A ᑌ B) = p(A) + p(B)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

A ᑎ B

A

“A intersection B”

Only elements that are in both A and B

When A ᑎ B = ∅, that means A and B don’t have any intersecting elements, hence they’re mutually exclusive

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

A ᑌ B

A

“A union B”

Contains all elements by A or B, or both.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

When representing the Probability Calculus in _______________ ___________, A and B are no longer “___________,” they become “_________________.”

A

propositional logic; Events; Propositions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

If Propositions ____ and ___ are ______________-______________ (meaning both can’t occur at the same time such that ___ ᐱ ___ = ____, AND that both ___ and ___ don’t have any __________ in common), then only one proposition must be _________, and the other must be _________.

A

A; B; mutually exclusive; A; B; ∅; A; B; elements; true; false

17
Q

Probability Calculus (Propositional Logic)

A

1) 0 ≥ p(A) ≥ 1

2) If proposition A is logically true (such that it’s impossible for A to be false), then p(A)=1.

3) If A ᐱ B = ∅, then p(A ᐯ B) = p(A) + p(B)

18
Q

The difference between 1st Set of Probability Calculus and Propositional Logic set of Probability Calculus

A

Probability calculus (on it’s own) assigns probabilities to events, THINGS THAT EXIST IN THE EXTERNAL WORLD.

Probability Calculus, formalized in propositional logic, assings probabilities to LINGUISTIC ENTITIES, aka. propositions.

19
Q

Theorem 6.1: p(A) + p(¬A) = 1

(Give steps on how to reach this answer)

A

p(A) + p(¬A) = 1

  • A ᐯ ¬A is Logical Truth.
  • Axiom 2 tells us, “If proposition A is a logical truth, then p(A) = 1
  • Therefore, p(A ᐯ ¬A) = 1
  • A and ¬A are Mutually Exclusive.
  • Axiom 3 tells us, “If A ᐱ B = ∅, then p(A ᐯ B) = p(A) + p(B).”
    -Therefore, p(A ᐯ ¬A) = p(A) + p(¬A)

Taken together, p(A) + p(¬A) = p(A ᐯ ¬A) = 1

20
Q

Theorem 6.2: If A and B are logically equivalent, then p(A) = p(B)

(Give steps on how to reach this answer)

A

If A and B are logically equivalent, meaning they the same TRUTH VALUE,
then A and ¬B are mutually exclusive.

  • Axiom 3: If A ᐱ B = ∅, then p(A ᐯ B) = p(A) + p(B).
  • So it follows, that p(A ᐯ ¬B) = p(A) + p(¬B)

The compound sentence A ᐯ ¬B is true for some logical reasons (it’s Logically True).

  • Axiom 2: If A is Logically True, then p(A) =1
  • So it follows, that p(A ᐯ ¬B) = 1

Taken together, if p(A ᐯ ¬B) = 1, and p(A ᐯ ¬B) = p(A) + p(¬B), then p(A) + p(¬B) = 1.

Were not done yet though! Now we need to refer back to Theorem 6.1: p(A) + (¬A) = 1.

Plugging this theorem in for our sentence ¬B, we get p(B) + p(¬B) = 1

p(A) + p(¬B) = 1
p(B) + p(¬B) = 1
In both of these equations, we yield the same value for ¬B (1), so it follows that p(A) = p(B)

21
Q

Theorem 6.3 i) p(A v B) = p(A) + p(B) - p(A ∧ B)

Theorem 6.3 ii) p(A → B) = p(¬A) + p(B) - p(¬A ∧ B)

A

Proof pt. i)
- p(A v B) and p(¬A ∧ B) v p(A ∧ ¬B) v p(A ∧ B) are logically equivalent.
- Great! Axiom 2: If proposition A is logically equivalent, then p(A)=1
- However, p(¬A ∧ B) and p(A ∧ ¬B) v p(A v B) are mutuallly exclusive.
- Okay then, Axiom 3: If A ∧ B = ∅, then p(A v B) = p(A) + p(B)
- p(A v B) = p( (p(¬A ∧ B) ) v (p(A ∧ ¬B) v p(A v B) ) = p(¬A ∧ B) + [p(A ∧ ¬B) v p(A v B)]
- *You can simplify the problem, the rightmost equations are equivalent with A.
- p(A v B) = p(¬A ∧ B) + p(A)
- *Now we need p(B) to complete equation at it’s stated in the 3rd axiom (p(A v B) = p(A) + p(B)), and we need to eliminate p(¬A ∧ B). Why don’t we try adding p(A ∧ B) to both sides!
- p(A v B) + p(A ∧ B) = p(¬A ∧ B) + p(A) + p(A ∧ B)
- *You can simplify the problem, ¬A ∧ B and A ∧ B are equivalent w/ B.
- p(A v B) + p(A ∧ B) = p(B) + p(A)
- p(A v B) = p(B) + p(A) - p(A ∧ B)

Proof pt. ii)
- Same as Proof pt. i). Just be sure to substitute A w/ ¬A such that…
p(¬A v B) = p(B) + p(¬A) - p(¬A ∧ B)

22
Q

Rule for Conjunctive Probability with Independent Conjuncts

A

p(A Ʌ B) = p(A) × p(B)

23
Q

In most decision problems, ______________ is _______________.

A

probability; conditional

24
Q

p(A|B) (First Definition of A|B )

A

p(A|B) = p(A ∧ B) / p(B)

25
Q

p(A|B) (Second Definiton of A|B)

A

A is indpendent of B iff. p(A) = p(A|B)

26
Q

Theorem 6.4 “Independence”

A

If A is independent of B, then the probability of p(A) = p(A|B) / p(B)

27
Q

Baye’s Theorem Definition

A
  • Calculates Conditional Probabilities
  • Shows how conditional probabilies are “reserved.”
28
Q

Theorem 6.5: Baye’s Theorem – “Inverse Probability Law”

A

p(B|A) = p(B) x p(A|B) / p(A)

29
Q

Theorem 6.6: Bayes Theorem

A

p(B|A) = p(B) x p(A|B) / [p(B) x p(A|B)] + [p(¬B) x p(A|¬B)]

30
Q

Prior Probability

A

The unconditional probabilty p(B) (in Bayes’ Theorem).

May very from case to case, and can’t be known until there’s a body of data necessary to calculate it.

31
Q

A _________ to use in problems invovling Bayes’ Theorem, is assigning real _____________ (___-___) to the __________________ (____________) _________________ [p(B)]. In doing this we can find, on the other side of the Bayes’ Theorem problem, the _____________ _______________.

[See Notes for Example]

A

trick; numbers; prior (unknown) probability; posterior probability.

32
Q

In a problem, if you’re given ______ conditional prior and ____ unconditional priors, you should use the ___________ version of Bayes’ Theorem. If you’re given ____ conditional priors and ______ unconditional prior, you should employ the ___________ _____________ version of Bayes’ Theorem.

A

1; 2; simple; 2; 1; expanded denominator

33
Q

If a problem asks you for the probability of an event, you should find the probability of ________ getting at least getting that event, such that we’renot finding ___, but __ - __. Provide an example…

A

NOT; p; 1-p

If you’re asked to find the probability of rolling a one, on a six-sided fair dice, after rolling 4 times, you should calculate the probability of NOT rolling a one, out of 4 tries: 1 - (5/6 * 5/6 * 5/6 * 5/6).

34
Q

REMEMBER: Use the ___-___ rule when there’s not a direct or _____________ way to get at probabilities. Were supposed to use this rule when we can identify ____________ cases very easily–more easily than we can identify ____________ cases.

A

1-p; convenient; negative; positive

ex. (You’re being asked after rolling a six-sided fair die twice, what’s the probability of the Sum of the two rolls is 4. The only combinations, out of 36 are ⟨1,3⟩ ⟨2,2⟩ ⟨3,1⟩
that you’ll get a sum of four. Therefore, the p(S) = 3/36 = 1/12 = 0.083. HERE, it was alot more convenient to find the positive cases. This is a case in which you absolutely should NOT use the 1 - p rule, because then you’d be finding the negative cases–33/36 negative cases to be exact–when it’s SO much easier to just find the positive cases.)