Scribd
Upload
5 views4 pages

03 AxiomProbability 0921

Uploaded by

Jack Lam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
5 views4 pages

03 AxiomProbability 0921

Uploaded by

Jack Lam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

NTHU MATH 2810, 2023 Lecture Notes

p. 3-6
 Example (Waiting for a success):
 Play roulette until a win.

 Ω = {1, 2, 3, …}.

 P = ??

 Example (Uniform Spinner):


 Random Angle (in radians).

 Ω = (−π, π].

 P= ??

• The Modern Approach


A probability measure on Ω is a function P from subsets of Ω to
the real number (or [0, 1]) that satisfies the following axioms:
(Ax1) Non-negativity. For any event A, P(A)≥0.
(Ax2) Total one. P(Ω)=1.
(Ax3) Additivity. If A1, A2, …, is a sequence of mutually
exclusive events, i.e., Ai∩Aj=∅ when i≠j, then
P (A1 ∪ A2 ∪ · · ·) = P (A1 ) + P (A2 ) + · · · .
p. 3-7
 Notes:
 These axioms restrict probabilities, but do not define them.
 Probability is a property of events.
Define Probability Measures in a Discrete Sample Space.
 Q: Is it required to define probabilities directly on every
events? (e.g., n possible outcomes in Ω, 2n−1 possible events)
 Suppose Ω = {ω1, ω2, …}, finite or countably infinite,
let satisfy
p(w) ≥ 0 for all ω ∈ Ω and ω∈Ω p(w) = 1.

 Let

for A⊂Ω, then P is a probability measure. (exercise)


(Q: how to define p?)
made by S.-W. Cheng (NTHU, Taiwan)
NTHU MATH 2810, 2023 Lecture Notes
p. 3-8
 Example: In the classical approach, p(ω) = 1/#Ω. For example,
throw a fair dice, Ω={1, …, 6}, p(1)=…=p(6)=1/6 and
P(odd)=P({1, 3, 5})=p(1)+p(3)+p(5)=3/6=1/2.
 Example (non equally-likely events): Throwing an unfair dice
might have p(1)=3/8, p(2)=p(3)=…=p(6)=1/8, and
P(odd)=P({1, 3, 5})=p(1)+p(3)+p(5)= 5/8. (c.f., Examples in
LNp.3-5)
 Example (Waiting for Success – Play Roulette Until a Win):
 Let r=9/19 and q=1−r=10/19

 Ω = {1, 2, 3, …}
 Intuitively, p(1)=r, p(2)=qr, p(3)=q2r, …, p(n)=qn−1r, … >0,
and ∞ ∞
r
p(n) = rq n−1 = = 1.
n=1 n=1
1−q
 For an event A⊂Ω, let
P (A) = n∈A p(n).

For example, Odd={1, 3, 5, 7, …}


∞ ∞ ∞ p. 3-9

P (Odd) = p(2k + 1) = rq (2k+1)−1 = r q 2k


k=0 k=0 k=0
= r/(1 − q 2 ) = 19/29.
 Reading: textbook, Sec 2.3 & 2.5

Some Consequences of the 3 Axioms


• Proposition: For any sample space Ω, the probability
of the empty set is zero, i.e.,

• Proposition: For any finite sequence of mutually exclusive events


A1, A2, …, An,

made by S.-W. Cheng (NTHU, Taiwan)


NTHU MATH 2810, 2023 Lecture Notes

• Proposition: If A is an event in a sample space Ω and Ac is the


p. 3-10

complement of A, then
P (Ac ) = 1 − P (A).

• Proposition: If A and B are events in a sample space Ω and A⊂B,


then P (A) P (B) and P (B − A) = P (B ∩ Ac ) = P (B) − P (A).

 Example (摘自“快思慢想”, Kahneman).


琳達是個三十一歲、未婚、有話直說的聰明女性。她主修
哲學，在學生時代非常關心歧視和社會公義的問題，也參
與過反核遊行。下面那一個比較可能？
 琳達是銀行行員。

 琳達是銀行行員，也是活躍的女性主義運動者。

p. 3-11

• Proposition: If A is an event in a sample space Ω, then


0 P (A) 1.

• Proposition: If A and B are two events in a sample space Ω, then

• Proposition: If A1, A2, …, An are events in a sample space Ω, then

made by S.-W. Cheng (NTHU, Taiwan)


NTHU MATH 2810, 2023 Lecture Notes
p. 3-12
• Proposition (inclusion-exclusion identity): If A1, A2, …, Anare any
n events, let Q: For an outcome w contained in
m out of the n events, how many
times is its probability p(w)
repetitively counted in σ1 , …, σn?

then

made by S.-W. Cheng (NTHU, Taiwan)

You might also like