Basic Probability What Every Math Student Should Know 2nd Edition H. C. Tijms 2024 scribd download
Basic Probability What Every Math Student Should Know 2nd Edition H. C. Tijms 2024 scribd download
https://ebookfinal.com/download/50-studies-every-plastic-surgeon-
should-know-c-scott-hultman/
https://ebookfinal.com/download/what-every-engineer-should-know-about-
decision-making-under-uncertainty-1st-edition-john-x-wang/
https://ebookfinal.com/download/what-every-engineer-should-know-about-
starting-a-high-tech-business-venture-1st-edition-eric-koester/
https://ebookfinal.com/download/21-candlesticks-every-trader-should-
know-1st-edition-melvin-pasternak/
The Golem at Large What You Should Know about Technology
2nd Edition Harry Collins
https://ebookfinal.com/download/the-golem-at-large-what-you-should-
know-about-technology-2nd-edition-harry-collins/
https://ebookfinal.com/download/network-marketing-what-you-should-
know-1st-edition-jeffrey-a-babener/
https://ebookfinal.com/download/the-language-of-
architecture-26-principles-every-architect-should-know-1st-edition-
andrea-simitch/
https://ebookfinal.com/download/50-landmark-papers-every-acute-care-
surgeon-should-know-1st-edition-stephen-m-cohn-editor/
https://ebookfinal.com/download/teaching-medicine-what-every-
inpatient-physician-needs-to-know-1st-edition-harrod/
Basic Probability What Every Math Student Should
Know 2nd Edition H. C. Tijms Digital Instant Download
Author(s): H. C. Tijms
ISBN(s): 9789811237492, 9811237492
Edition: 2nd
File Details: PDF, 10.10 MB
Year: 2021
Language: english
12295_9789811237492_tp.indd 1 24/6/21 11:28 AM
B1948 Governing Asia
For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance
Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy
is not required from the publisher.
Printed in Singapore
Preface
v
April 9, 2021 18:9 book-9x6 12295-main page vi
vi Preface
Contents
Preface v
vii
April 9, 2021 18:9 book-9x6 12295-main page viii
viii Contents
Index 173
April 9, 2021 18:9 book-9x6 12295-main page 1
Chapter 1
Combinatorics and Calculus for Probability
1
April 9, 2021 18:9 book-9x6 12295-main page 2
How many different ways can you order a number of different objects
such as letters or numbers? For example, what is the number of
different ways that the three letters A, B, and C can be ordered?
By writing out all the possibilities ABC, ACB, BAC, BCA, CAB,
and CBA, you can see that the total number is 6. This brute-force
method of writing down all the possibilities and counting them is
naturally not practical when the number of possibilities gets large,
as is the case for the number of possible orderings of the 26 letters
of the alphabet. You can also determine that the three letters A,
B, and C can be ordered in 6 different ways by reasoning as follows.
For the first position, there are 3 available letters to choose from, for
the second position there are 2 letters left over to choose from, and
only one letter for the third position. Therefore the total number of
possibilities is 3 × 2 × 1 = 6. The general rule should now be evident.
Suppose that you have n distinguishable objects. How many ordered
arrangements of these objects are possible? Any ordered sequence
of the objects is called a permutation. Reasoning in the same way
as above gives that there are n ways for choosing the first object,
leaving n − 1 choices for the second object, etc. Therefore the total
number of ways to order n distinguishable objects is equal to the
product n × (n − 1) × · · · × 2 × 1. This product is denoted by n! and
is called ‘n factorial’. Thus, for any positive integer n,
n! = 1 × 2 × · · · × (n − 1) × n.
A useful convention is
0! = 1,
which simplifies the presentation of several formulas to be given be-
low. Note that n! = n × (n − 1)! and so n! grows very quickly as
n gets larger. For example, 5! = 120, 10! = 3 628 800 and 15! =
1 307 674 368 000. Summarizing, for any positive integer n,
the total number of ordered sequences (permuta-
tions) of n distinguishable objects is n!.
April 9, 2021 18:9 book-9x6 12295-main page 3
Armed with this knowledge, you are asked to argue that in the lottery
6/45 the six lotto numbers, which are drawn one by one from 1 to
45, will appear in either an ascending or descending order with a
1
probability of 360 .
An interesting question is how many different words, whether or not
existing, can be constructed from a given number of letters where
some letters appear multiple times. For example, how many different
words can be constructed from five letters A, two letters B, two
letters R, one letter C, and one letter D? To answer this question,
imagine that the five letters A are labeled as A1 to A5 , the two letters
B as B1 and B2 , and the two letters R as R1 and R2 . Then you have
11 different letters and the number of ways to order those letters is
11!. The five letters A1 to A5 , the two letters B1 and B2 , and the
two letters R1 and R2 can among themselves be ordered in 5!×2!×2!
ways. Each of these orderings gives the same word if you replace A1
to A5 by A, B1 and B2 by B, and R1 and R2 by R. Thus the total
number of different words that can be formed from the original 11
letters is
11!
= 83 160.
5! × 2! × 2!
Thus, if you thoroughly mix the eleven letters and then put them in
a row, the probability of getting the word ABRACADABRA is equal
to 83 1160 .
How many different juries of three persons can be formed from five
persons A, B, C, D, and E? By direct enumeration you see that the
answer is 10: {A, B, C}, {A, B, D}, {A, B, E}, {A, C, D}, {A, C, E},
{A, D, E}, {B, C, D}, {B, C, E}, {B, D, E}, and {C, D, E}. In this
problem, the order in which the jury members are chosen is not rele-
vant. The answer 10 juries could also have been obtained by a basic
principle of counting. First, count how many juries of three persons
are possible when attention is paid to the order. Then determine
how often each group of three persons has been counted. Thus the
reasoning is as follows. There are 5 ways to select the first jury mem-
ber, 4 ways to then select the next member, and 3 ways to select the
April 9, 2021 18:9 book-9x6 12295-main page 4
5×4×3×2×1 5!
= .
3! × 2! 3! × 2!
In general, you can calculate that the total number of possible ways
to choose a jury of k persons out of a group of n persons is equal to
dex
= ex for all x.
dx
Take for granted that c = limh→0 (ah − 1)/h exists. Thus a0 (x) = ca(x).
When is the constant c = 1? The answer is if a = e. To see this, note that
the condition limh→0 (ah − 1)/h = 1 is the same as a = limh→0 (1 + h)1/h =
limn→∞ (1 + n1 )n = e. This shows that a0 (x) = a(x) if a = e. A more general
result is that f (x) = ex is the only function satisfying the differential equation
f 0 (x) = f (x) with the boundary condition f (0) = 1.
x2 x3
ex = 1 + x + + + ··· for all x.
2! 3!
The sought probability that two or more people share a same birth-
day is one minus the probability that each person has a different
birthday. Thus
Natural logarithm
d ln(y) 1
= for y > 0.
dy y
April 9, 2021 18:9 book-9x6 12295-main page 10
1
1 + x + x2 + · · · = for |x| < 1,
1−x
or, shortly, ∞ k 1
P
k=0 x = 1−x for |x| < 1. You can easily verify this
result by working out (1 − x)(1 + x + x2 + · · · + xm ) as 1 − xm+1 .
If you take |x| < 1 and let m tend to infinity, then xm+1 tends to 0.
This gives (1 − x)(1 + x + x2 + · · · ) = 1 for |x| < 1, which verifies
the desired result. Differentiating the geometric series term by term
1 1
and noting that 1−x has (1−x)2 as derivative, you get
1
1 + 2x + 3x2 + · · · = for |x| < 1
(1 − x)2
P∞ k−1 = 1
or, shortly, k=1 kx (1−x)2
for |x| < 1. Similarly, you get
P∞ k−2 2
k=2 k(k−1)x = (1−x)3 for |x| < 1. For the geometric probability
model to be met in the next chapters, these formulas lead to
∞ ∞
X 1 X 2−p
k(1 − p)k−1 p = and k 2 (1 − p)k−1 p = .
p p2
k=1 k=1
Chapter 2
Basics of Probability
11
April 9, 2021 18:9 book-9x6 12295-main page 12
Basics of Probability 13
Basics of Probability 15
blue die, and j represents the number of points rolled with the red
die. The dice are fair so that an appropriate probability model is
1
constructed by assigning the same probability 36 to each element of
the sample space. This model is an instance of the Laplace model:
in order to find the probability of an event, you count the number of
favorable outcomes and divide them by the total number of possible
outcomes. Let A be the event that the largest number rolled is 4.
The event A occurs only if the experiment gives one of the seven
outcomes (1, 4), (2, 4), (3, 4), (4, 4), (4, 3), (4, 2) and (4, 1). Hence
7
P (A) =
36
gives the probability of getting 4 as largest number in one roll of two
fair dice.
Sample points may be easily incorrectly counted. In his book Opera
Omnia the German mathematician Gottfried Wilhelm Leibniz (1646–
1716) – inventor of differential and integral calculus along with Isaac
Newton – made a famous mistake by stating: “with two dice, it is
equally likely to roll twelve points than to roll eleven points, because
one or the other can be done in only one manner”. He argued: two
sixes for a sum 12, and a five and a six for a sum 11. However, there
are two ways to get a sum 11, as is obvious by imagining that one
die is blue and the other is red. Alternatively, you may think of two
rolls of a single die rather than a single roll of two dice.
Another psychologically tempting mistake that is sometimes made
is to treat sample points as equally likely while this is actually not
the case. This mistake can be illustrated with a famous misstep of
Jean le Rond d’Alembert (1717–1783), who was one of the foremost
intellectuals of his time. D’Alembert made the error to state that
the probability of getting heads in no more than two coin tosses is 32
rather than 34 . He reasoned as follows: “once heads appears upon the
first toss, there is no need for a second toss. The possible outcomes of
the game are thus H, T H and T T , and so the required probability is
2
3 ”. However, these three outcomes are not equally likely, but should
be assigned the respective probabilities 12 , 14 and 14 . The correct
answer is 43 , as would be immediately clear from the sample space
{HH, HT, T H, T T } for the experiment of two coin tosses.
April 9, 2021 18:9 book-9x6 12295-main page 16
5 5 1 5 i−1 1
pi = × ··· × × = × for i = 1, 2, . . . .
6 6 6 6 6
Let A be the event that desperado A fires the fatal shot. This event
occurs
P∞ for the outcomes 1, 3, 5, . . .. Thus, using the geometric series
k 1
k=0 x = 1−x for |x| < 1, you get
∞ ∞
X 1 X 25 k 1 1
P (A) = p2k+1 = = × ,
6 36 6 1 − 25/36
k=0 k=0
6
and so the probability that desperado A will fire the fatal shot is 11 .
Basics of Probability 17
d/2
d/2 d/2 a
d/2
(a − d)2
P (the coin will fall entirely within a square) = .
a2
Complement rule
Basics of Probability 19
could obtain a six in four rolls or less of a single die and the bet that
he could obtain a double six with two dice in 24 rolls or less. The
Chevalier believed that the chance of winning the bet was the same
in both games (can you explain why the respective chances are not
4 24
6 and 36 ?). In reality, however, he won the first game more often
than not. The Chevalier approached the mathematician Blaise Pas-
cal for clarification. This inquiry led to a correspondence between
the two famous French mathematicians Blaise Pascal (1623–1662)
and Pierre de Fermat (1601–1665).5 They mathematically clarified
the dice problem by simply calculating the chances of not rolling a
six or double six, see Problem 2.2 below.
The complement rule says that, for any event A,
P (A) = 1 − P (A),
Problem 2.4. You are randomly dealt four cards from a deck of 52
playing cards with four aces. What is the probability of getting at
least one ace? (answer: 0.2813)
Problem 2.5. You have two gift cards, each loaded with 10 free
drinks from your favorite coffee shop. Each time you get a drink,
you randomly pick one of the cards to pay with. One day it happens
for the first time that the waiter can’t accept the card because it
does not have any drink credits left on it. What is the probability
that the other card has also no free drinks on it? (answer: 0.1762)
April 9, 2021 18:9 book-9x6 12295-main page 21
Basics of Probability 21
Problem 2.6. In a game show, a father and his daughter are stand-
ing in front of three closed doors, behind which, a car, the key to the
car, and a goat are hidden in random order. Each of them can open
up to two doors, one at a time, and this must be done out of sight
of the other. The daughter is given the task of finding the car, and
the father must find the key. Only if both are successful, they get to
keep the car. Father and daughter are allowed to discuss a strategy
before the game starts. What is an optimal strategy? What is the
maximum probability of winning the car? (answer: 46 )
Problem 2.9. Two people have agreed to meet at the train station.
Independently of one other, each person is to appear at a random
moment between 12 p.m and 1 p.m. What is the probability that
they will meet within 10 minutes of each other? (answer: 11
36 )
4 3
P (the first two cards are aces) = × .
52 51
April 9, 2021 18:9 book-9x6 12295-main page 22
where P (A and B) stands for the probability that both event A (‘the
first card is an ace’) and event B (‘the second card is an ace’) will
occur, P (B | A) is the notation for the conditional probability that
event B will occur given that event A has occurred.6 In words, the
unconditional probability that both event A and event B will occur is
equal to the unconditional probability that event A will occur times
the conditional probability that event B will occur given that event
A has occurred. This is one of the most useful rules in probability.
Basics of Probability 23
1 128
P (no apartment is vacant on the top floor) = = 0.7325.
1 540
Example 2.6. Three boys and three girls are planning a dinner
party. They agree that two of them will do the washing up, and they
draw lots to determine which two it will be. What is the probability
that two boys will wind up doing the washing up?
Example 2.7. What is the probability that you must pick five or
more cards from a shuffled deck of 52 cards before getting an ace?
which gives 49
P
k=5 pk = 0.7187. It never hurts to solve a problem in
different ways. It allows you to double check your answer.
The foregoing examples show that when you use an approach based
on conditional probabilities to solve the problem, you usually go
straight to work without first defining a sample space. The counting
approach, however, does require the specification of a sample space.
If both approaches are possible for a given problem, then the ap-
proach based on conditional probabilities will, in general, be simpler
than the counting approach.
For events A and B with nonzero probabilities, the formula
P (A and B)
P (B | A) =
P (A)
April 9, 2021 18:9 book-9x6 12295-main page 25
Basics of Probability 25
quantifies how the original probability P (B) changes when new in-
formation becomes available. If P (B | A) = P (B), then the events
A and B are said to be independent. An equivalent definition of
independence is P (A and B) = P (A)P (B). The concept of indepen-
dence will be further explored in Section 2.7.
Beginning students sometimes think that independent events A
and B with nonzero probabilities are disjoint. This is not true. The
explanation is that P (A and B) = 0 if A and B are disjoint, whereas
P (A and B) = P (A)P (B) > 0 if A and B are independent.
Problem 2.12. Someone has rolled two dice out of your sight. You
ask this person to answer “yes or no” on the question whether there
is a six among the two rolls. He truthfully answers “yes.” What is
1
the probability that two sixes have been rolled? (answer: 11 )
Problem 2.15. Four British teams are among the eight teams that
have reached the quarter-finals of the Champions League soccer.
What is the probability that the four British teams will avoid each
other in the quarter-finals draw if the eight teams are paired ran-
8
domly? (answer: 35 ) Hint: think of a bowl containing four red and
four blue balls, where you remove each time two randomly chosen
balls from the bowl. What is the probability that you remove each
time a red and a blue ball? Solving a probability problem becomes
often simpler by casting the problem into an equivalent form.
Problem 2.16. If you pick at random two children from the Johnson
family, the chances are 50% that both children have blue eyes. How
many children does the Johnson family have and how many of them
have blue eyes? (answer: 4 and 3)
Suppose that a closed box contains one ball. This ball is white. An
extra ball is added to the box and the added ball is white or red
with equal chances. Next one ball is blindly removed from the box.
What is the probability that the removed ball is white? A natural
reasoning is as follows. The probability of removing a white ball is
1 if a white ball has been added to the box and is 12 if a red ball has
been added to the box. It is intuitively reasonable to average these
conditional probabilities over the probability that a white ball has
been added and the probability that a red ball has been added. The
latter two probabilities are both equal to 12 . Therefore
1 1 1 3
P (the removed ball is white) = 1 × + × = .
2 2 2 4
This is an application of the law of conditional probability. This law
calculates a probability P (A) with the help of appropriately chosen
April 9, 2021 18:9 book-9x6 12295-main page 27
Basics of Probability 27
Basics of Probability 29
P (B)P (A | B)
P (B | A) =
P (A)
for any two events A and B with P (A) > 0. The derivation of this
formula is strikingly simple. The basic form of Bayes’ rule follows
directly from the definition of conditional probability:
P (B and A) P (B)P (A | B)
P (B | A) = = .
P (A) P (A)
o(G)
p= .
1 + o(G)
P (H | E) P (H) P (E | H)
= × .
P (H | E) P (H) P (E | H)
What does this formula say and how to use it? This is easiest ex-
plained with the help of an example.
9
This rule is obtained as follows. The basic form of the formula of Bayes gives
that P (H | E) = P (H)P (E | H)/P (E) and P (H | E) = P (H)P (E | H)/P (E).
Taking the ratio of these two expressions, P (E) cancels out and you get Bayes’
rule in odds form. The derivation shows that the formula is also true when H
would not be the complement of H. That is, for any two hypotheses H1 and H2 ,
P (H1 |E) P (H1 )
the general Bayes formula P (H2 |E)
= P (H2 )
×P (E|H1 )
P (E|H2 )
applies.
April 9, 2021 18:9 book-9x6 12295-main page 31
Basics of Probability 31
Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.
ebookfinal.com