You are on page 1of 46

Elementary Probability Theory, Fourth Edition

Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

Probability Theory and Statistics


Zhengyao Bai, PhD, Professor
School of Information Science & Engineering, Yunnan University

Email: baizhy@ynu.edu.cn
Mobile: 13577060089
QQ & Wechat: 1302255793
Room: 1417#, SISE Building

0
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Basic formulas
• Sequential sampling
• Polya's urn scheme
• Independent and relevance

1
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• The probability of a set A is its weighted proportion
relative to the sample space Ω: P(A) = |A|/|Ω|.

• When Ω is countable and each point ω has the weight


P(ω) = P({ω}) attached to it:

2
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• In many questions we are interested in the proportional
weight of one set A relative to another set S. this means
the proportional weight of the part of A in S, namely the
intersection A ∩ S, or AS, relative to S.

It is called the conditional probability of A relative to S.

3
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Example 1. All students on a certain college campus are
polled as to their reaction to a certain presidential
candidate.
Let D denote those who favor him. Now the student population Ω
may be cross-classified in various ways, for instance according to sex,
age, race, etc. Let A = female, B = black, C = of voting age.
Then Ω is partitioned into 8 subdivisions ABC, ABCc,..., AcBcCc. The
set D will in general cut across the various divisions.

These two formulas denote the proportion of male black students of


voting age who favor the candidate, and the proportion of male students
of voting age who do not favor the candidate, respectively.
4
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Example 2. A perfect die is thrown twice. Given that the
total obtained is 7, what is the probability that the first
point obtained is k, 1 ≤ k ≤ 6?
Let us look at the outcome space: 62 points in the space

5
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Example 2. The outcomes with total equal to 7 are those
on the second diagonal, and their number is 6. Hence the
conditional probability is equal to 1/6.

6
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Example 2. Let X1 and X2 respectively denote the point
obtained in the first and second throw.

Other examples:

7
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Example 2. Given X1 = 4, what is the probability that X2
= k provided that the two trials are independent?

Since the second throw is not affected by the first, so the


conditional probability P{X2 = k | X1 = 4} must be the same
as the unconditional one P{X2 = k}.

8
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Example 3. Suppose you toss a biased coin repeatedly
until a head turns up. Let X be the waiting time.
Knowing that it has fallen tails three times, what is the
probability that it will fall heads within the next two
trials?
It is the conditional probability:

How to compute the probability?

9
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Example 3. We know that

From which we can calculate: How to get the sum?

10
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Example 3. The probability that the coin falls heads (at
least once in two trials):

11
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Example 4. An analogy between the geometrical
distribution and the exponential distribution:

waiting time
12
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Example 5. Consider all families with two children and
assume that boys and girls are equally likely. The sample
space may be denoted schematically as: {(bb), (bg), (gb),
(gg)}. The order in each pair is the order of birth. the 4
points have probability 1/4 each.

13
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Example 5. Q1: If a family is chosen at random from Ω
and found to have a boy in it, what is the probability that
it has another boy, namely that it is of the type (b, b)?

14
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Examples of conditioning
• Example 5. Q2: If a child is chosen at random from these
families and is found to be a boy, what is the probability
that the other child in his family is also a boy?

15
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Simple but fundamental propositions involving
conditional probabilities
• Proposition 1 (general mulplicative formula) For
arbitrary events A1, A2, ..., An, we have

provided P(A1A2...An-1) > 0.


• Application of multipllicative formula

16
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Proposition 2 (total probability formula) Sppose that

is a partition of the sample space into disjoint sets. Then


for any set B we have

17
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Proposition 3 (Bayes' Theorem) Under the assumption
and notation of Proposition 2, we have also

provided P(B) > 0. The denominator above is equal to


P(B) by Proposition 2.

18
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Example 6. What is the probability of throwing six
perfect die and getting six different faces? Put

etc. Assuming that the dice act independently, we have

(Proposition 1)
19
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Example 7. The family dog is missing after the picnic.
Three hypotheses are suggested:
(A) it has gone home;
(B) it is still worrying that big bone in the picnic area;
(C) it has wandered off into the woods.
The a priori probabilities, which are assessed from the
habits of the dog, are estimated respectively to be 1/4 , 1/2 ,
1/4 . A child each is sent back to the picnic ground and the
edge of the woods to look for the dog. If it is in the former
area, it is a cinch (90%) that it will be found; if it is in the
latter, the chance is only a toss-up (50%).
20
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Q1: What is the probability that the dogwill be found in
the park?
Let D = “dog will be found in the park.” Then we have

21
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Q2: What is the probability that the dog will be found at
home?
Let D' = “the dog will be found at home.” and assume
that P(D'| A) = 1. Clearly P(D'| B) = P(D'| C) = 0. Then we
have

22
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Q3: What is the probability that the dog is “lost”? It is

23
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Example 8. Urn one contains 2 black and 3 red balls; urn
two contains 3 black and 2 red balls. We toss an unbiased
coin to decide on the urn to draw from but we do not
know which is which. Suppose the first ball drawn is
black and it is put back.

U1 U2

24
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Example 8. Q1: What is the probability that the second
ball drawn from the same urn is also black?
For two urns U 1 and U 2 , the a priori probability that
either one is chosen by the coin-tossing is 1/2:

Denote the event that the first ball is black by B1, that the
second ball is black by B2.

25
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Example 8. Q1: What is the probability that the second
ball drawn from the same urn is also black?
Let A1 = B1 is from U1, A2 = B1 is from U2. The second
drawing is conditioned on B 1 , the probabilities of the
hypotheses are really conditional ones:

it is obvious that

26
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Example 8. Q1: What is the probability that the second
ball drawn from the same urn is also black?
We obtain the conditional probability:

comparing this with

We see that the knowledge of the first ball drawn being


black has strengthened the probability of drawing a second
black ball, because it has increased the likelihood that we
have picked the urn with more black balls.
27
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Basic formulas
• Example 8. Q2: Given that the first two balls drawn are
both black and put back, what is the probability of
drawing a third black ball from the same urn?
In similar notations, we have:

It is easy to see that we can extend the result to any


number of drawings:

28
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Sequential sampling
• An urn contains b black balls and r red balls. One ball is
drawn at a time without replacement. Let X n = 1 or 0
depending on whether the nth ball drawn is black or red.
Each sample point ω is then just the sequence {X1(ω),
X2(ω),... , Xb+r(ω)}, briefly {Xn, 1 ≤ n ≤ b + r}.

• Problem. A ball is drawn from the urn and discarded.


Without knowing its color, what is the probability that a
second ball drawn is black?

29
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Sequential sampling
• Let , then we have

30
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Sequential sampling
• P(B 2 ) = P(B 1 ), namely if we take into account both
possibilities for the color of the first ball, then the
probabilities for the second ball are the same as if no ball
had been drawn (and left out) before.

• Theorem 1. For each n, we have

31
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Sequential sampling
• Theorem 2 (Poisson’s Theorem). Suppose in an urn
containing b black and r red balls, n balls have been
drawn first and discarded without their colors being
noted. If m balls are drawn next, the probability that
there are k black balls among them is the same as if we
had drawn these m balls at the outset [without having
discarded the n balls previously drawn].

32
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Polya's urn scheme


• As before the urn contains b black and r red balls to
begin with, but after a ball is drawn each time, it is
returned to the urn and c balls of the same color are
added to the urn, where c is an integer, and when c < 0
adding c balls means subtracting -c balls.
• Polya’s urn model when c >= -1.

33
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Polya's urn scheme


• Problem. What is the probability that in Polya’s model
the first three balls drawn have colors {b, b, r} in this
order? or {b, r, b}? or {r, b, b}?

34
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Polya's urn scheme


• Theorem 3. The probability of drawing (from the
beginning) any specified sequence of k black balls and n
− k red balls is equal to

for all n ≥ 1 if c ≥ 0; and for 0 ≤ n ≤ b + r if c = −1.

35
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Independent and relevance


• An extreme and extremely important case of conditioning
occurs when the condition has no effect on the probability.

• Definition of Independent Random Variables. The


countably valued random variables X1, ... , Xn are said to
be independent iff for any real numbers x 1 , ... , x n , we
have

36
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Independent and relevance


• Proposition 4. We have for arbitrary countable sets S1, ... ,
Sn:

• Proposition 5. The events {X1 ∈ S1},... , {Xn ∈ Sn}are


independent.

• Proposition 6. Let 1, ... , n be an arbitrary real-valued


function on (−∞, ∞); then the random variables
1(X1), ... , n(Xn)
are independent.
37
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Independent and relevance


• Example 10. A letter from Pascal to Fermat (dated
Wednesday, 29th July, 1654), contains, among many other
mathematical problems, the following passage:

38
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Independent and relevance


M. de Mere told me that he had found a fallacy in the
theory of numbers, for this reason: If one undertakes to get
a six with one die, the advantage in getting it in 4 throws is
as 671 is to 625. If one undertakes to throw 2 sixes with two
dice, there is a disadvantage in undertaking it in 24 throws.
And nevertheless 24 is to 36 (which is the number of
pairings of the faces of two dice) as 4 is to 6 (which is the
number of faces of one die). This is what made him so
indignant and made him say to one and all that the
propositions were not consistent and Arithmetic was self-
contradictory: but you will very easily see that what I say is
correct, understanding the principles as you do.

39
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Independent and relevance


• Example 10. To throw a six with 1 die in 4 throws means
to obtain the point “six” at least once in 4 trials. Define Xn,
1 ≤ n ≤ 4, as follows:

• assume that X1, X2, X3, X4 are independent. Put An = {Xn =


6}; then the event in question is A1∪A2 ∪A3∪A4. It is
easier to calculate the probability of its complement. The
trials are assumed to be independent and the dice
unbiased.

40
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Independent and relevance


• Example 10. Next consider two dice; let (X'n, X''n) denote
the outcome obtained in the nth throw of the pair, and let:

Then

41
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Independent and relevance


• Example 11. If two points are picked at random from the
interval [0, 1],what is the probability that the distance
between them is less than 1/2?

The double integral can be easily


computed geometrically.

See the Figure. The probability


is equal to the area of S:

42
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Independent and relevance


• Example 12. Suppose X 1 , X 2 , ... , X n are independent
random variables with distributions F1, F2, ... , Fn. Let

Find the distribution functions of M and m.

Using (5.5.7), we have for each x

43
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter 5 Conditioning and Independence

• Independent and relevance


• Example 12. As for the minimum, it is convenient to
introduce the “tail distribution” Gj corresponding to each
Fj as follows:

Then we have, using the analogue of (5.5.2) this time


with Sj = (xj ,∞),

44
Elementary Probability Theory, Fourth Edition
Kai Lai Chung, Farid AitSahlia

Chapter
Chapter 6 5 Conditioning and Independence

To be updated…

THANKS
2020-10-7

You might also like