2024 Fall, STA 1005ab, AUBG

Review problems for the Midterm Exam

HI

October 20, 2024

Contents
1 Introduction to probability 2
1.1 Sample spaces and probability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Other basic rules about probability . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Conditional probability and independence of events . . . . . . . . . . . . . . . . . 6
1.4 Counting rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4.1 The mn-rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4.2 The extended mn-rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4.3 Permutations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.4.4 Permutations of non-distinct objects . . . . . . . . . . . . . . . . . . . . . . 10
1.4.5 Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.5 Summary of the counting rules and more problems . . . . . . . . . . . . . . . . . 12

2 Additional problems on probability 14

3 Probability Distributions 15
3.1 Mean, variance, standard deviation, and expectation . . . . . . . . . . . . . . . . . 18
3.2 Computing expected values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

4 Several discrete probability distributions 21

4.1 The Binomial distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.2 The Poisson Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.3 Approximating Binomial distribution with Poisson distribution . . . . . . . . . . 25

5 Continuous distributions 28
5.1 Basics on continuous distributions . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
5.2 The Normal distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
5.3 Applications of the normal distribution . . . . . . . . . . . . . . . . . . . . . . . . 36

1
1 Introduction to probability
1.1 Sample spaces and probability
The sample space is the set of all possible outcomes of a probability experiment. We denote it by
S. An event A consists of outcomes of the probability experiment, i.e., A is a subset of S.
One can think of the probability as a function P that to an event A assigns a number between
0 and 1. If S consists of finitely many outcomes, say S = {E1 , E2 , . . . , En }, then

P : S → [0, 1]
A 7→ P (A) .

A few facts:

1. The sum of the probabilities of all the outcomes in a sample space is 1; if the sample space
consists of finitely many outcomes, i.e., S = {E1 , E2 , . . . , En }, this means that

P (E1 ) + P (E2 ) + · · · + P (En ) = 1 .

2. The probability of any event A is a number (either a fraction or decimal) between and
including 0 and 1. This is denoted by 0 ≤ P (A) ≤ 1.

3. If an event D cannot occur (i.e., the event contains no members in the sample space), its
probability is 0.

4. If an event C is certain, then the probability of C is 1.

The most basic rule is that if A consists of (distinct) outcomes Ei1 , Ei2 , . . . , Eik , then

P (A) = P (Ei1 ) + P (Ei2 ) + · · · + P (Eik ) .

Problem 1. Two cold tablets are accidentally placed in a box containing two aspirin tablets. The four
tablets are identical in appearance. One tablet is selected at random from the box and is swallowed by
the first patient. A tablet is then selected at random from the three remaining tablets and is swallowed
by the second patient. Define the following events as specific collections of simple events:

(a) The sample space S

(b) The event A that the first patient obtained a cold tablet

(c) The event B that exactly one of the two patients obtained a cold tablet

(d) The event C that neither patient obtained a cold tablet

Solution: Let’s identify the event cold tablet is selected by C and the event aspirin tablet is selected
by A. Then

(a) The sample space S is S = {(CC), (CA), (AC), (AA)}.

(b) The event A that the first patient obtained a cold tablet, consists of the simple events
A = {(CC), (CA)}.

2
 (c) The event B that exactly one of the two patients obtained a cold tablet, consists of the
simple events B = {(CA), (AC)}.

(d) The event C that neither patient obtained a cold tablet, consists of just one simple event
C = {(AA)}.

Problem 2. Find the sample space for the gender of the children if a family has three children. Take into
account the birth order (and assume no twins). Use B for boy and G for girl. Assign probabilities to the
simple events in the sample space. What is the probability that a randomly selected family has exactly
two boys? At most two girls? Exactly three girls?

Solution: The sample space consists of

S = {E1 = (BBB), E2 = (BBG), E3 = (BGB), E4 = (BGG), E5 = (GBB),

E6 = (GBG), E7 = (GGB), E8 = (GGG)} .
It is natural to assign equal probabilities to these events, thus P (Ei ) = 1/8 for all i = 1, . . . , 8.
Denote the event A : exactly two boys, let B be the event at most two girls, and by C the event
exactly three girls. Then A = {E2 , E3 , E5 }, hence P (A) = P (E2 ) + P (E3 ) + P (E5 ) = 3/5. Sim-
ilarly, B = {E1 , E2 , E3 , E4 , E5 , E6 , E7 }, so P (B) = 7/8, and C = {E8 } hence P (C) = 1/8. To
summarize,

the probability for exactly two boys is P (A) = 3/8;

the probability for at most two girls is P (B) = 7/8;

the probability for exactly three girls P (C) = 1/8.

Problem 3. The game of roulette uses a wheel containing 38 pockets. Thirty-six pockets are numbered
1, 2, . . . , 36, and the remaining two are marked 0 and 00. The wheel is spun, and a pocket is identified as
the ”winner”. Assume that the observance of any one pocket is just as likely as any other.

(a) Identify the simple events in a single spin of the roulette wheel.

(b) Assign probabilities to the simple events.

(c) Let A be the event that you observe either a 0 or a 00. List the simple events in the event A and
find P (A).

(d) Suppose you placed bets on the numbers 1 through 18. What is the probability that one of your
numbers is the winner?

Solution: We denote by S the sample space.

(a) We identify the simple events as S = {E0 = 0, E1 = 1, . . . , E36 = 36, E37 = 00}.

(b) Each of these events is equally likely to occur as any other, so we should assign probability
1
to each of as there are exactly 38 possible outcomes.
38
3
 (c) If A is the event observe either a 0 or a 00, then A = {E0 , E37 }, and P (A) = P (E0 )+P (E37 ) =
2 1
= .
38 19
(d) Let B be the event: a number between 1 and 18 (inclusive) is drawn. Then B = {E1 = 1, E2 =
2, . . . , E18 = 18}. Then the probability for B is
18 9
P (B) = P (E1 ) + · · · + P (E18 ) = = .
38 19

Problem 4. A bowl with apples contains 1 red, 2 yellow and 3 green apples. A hungry student is about
to pick and eat two apples at random (the order of selection is not important).

(a) Display all simple events in the sample space.

(b) Assign probabilities to the simple events.

(c) What is the probability that he eats two apples of the same color?

Solution: We denote by S the sample space, by R red apple, by Y yellow apple, and by G
green apple.

(a) S = {E1 = (RY ), E2 = (RG), E3 = (Y Y ), E4 = (Y G), E5 = (GG)}.

(b) A pair of two apples can be selected from a set of 6 apples in N = 6 × 5/2 = 15 different
ways. In the pair with one red apple, R is selected in a unique way, while Y could chosen
in two different ways (think of the yellow ones being Y1 and Y2 ), and G can chosen in
three different ways (think of the greens being G1 , G2 and G3 ). Therefore, we should
2 3 1 6 3
have P (E1 ) = and P (E2 ) = . Similarly, P (E3 ) = , P (E4 ) = and P (E5 ) = .
15 15 15 15 15
Remark that we sure get
2+3+1+6+3
P (E1 ) + · · · + P (E5 ) = = 1.
15

(c) The event A : he eats two apples of the same color happens if he eats two yellow or two
green apples, i.e. A = {E3 , E5 }. Hence
1 3 4
P (A) = P (E3 ) + P (E5 ) = + = .
15 15 15

1.2 Other basic rules about probability

Let S be the sample space and A ⊂ S be an event. The set that consists of all elements of S that
are not elements of A is called the complement of A and we denote it by Ac or Ā.
Given events A, B ⊂ S we have

• (Ac )c ≡ A;

4
 • If A ⊂ B then Ac ⊃ B c ;

• If A = ∅ then Ac = S;

• For every set (event) A we have A ∩ A = A.

For any two events A and B, the probability that A or B will occur is

P (A or B) = P (A) + P (B) − P (A and B) .

We write this also as P (A ∪ B) = P (A) + P (B) − P (A ∩ B). It is known as the inclusion-exclusion

formula. For three events it reads as

P (A ∪ B ∪ C) = P (A) + P (B) + P (C) − P (A ∩ B) − P (A ∩ C) − P (B ∩ C) + P (A ∩ B ∩ C).

Problem 5. If one card is drawn from an ordinary deck of cards, find the probability of getting each
event:

(a) A spade or a queen or a king

(b) A club or a face card

(c) An ace or a diamond or a heart

(d) A 9 or a 10 or a spade or a club

Solution: Easy!
Problem 6. A jar contains four coins: a nickel (= 5 ¢), a dime (= 10 ¢), a quarter (= 25 ¢), and a
half-dollar. Three coins are randomly selected from the jar.

(a) List the simple events in S.

(b) What is the probability that the selection will contain the half-dollar?

(c) What is the probability that the total amount drawn will equal 60 ¢, or more?

Solution:

(a) S = {E1 = (5, 10, 25), E2 = (5, 10, 50), E3 = (5, 25, 50), E4 = (10, 25, 50)}.

(b) The simple events are equally likely. Let A50 be the event ’the selection contains the half-
dollar’. Then A = {E2 , E3 , E4 }, nA50 = 3, N = 4 and
nA50 3
P (A50 ) = = = 0.75 .
N 4

(c) The total total amount drawn will be less than 60 ¢ only if the triple E1 = (5, 10, 25) is
drawn. In all other cases the selection will include the half-dollar, and the total amount
will be at least 65. Thus

P ( the total amount drawn is ≥ 60 ¢) = P (A50 ) = .75 .

5
1.3 Conditional probability and independence of events
The conditional probability of event A, given that event B has occurred is

P (A ∩ B)
P (A|B) = , provided that P (B) ̸= 0 .
P (B)

Example 7. Consider a standard deck of 52 cards and draw a card randomly.

(a) What is the probability to draw a King?

(b) If you know that a face card has been drawn, what is the probability that it is a King?

Solution: Denote by A the event “King has been drawn”, and by B the event “face card has
been drawn”.
4 1
(a) To draw a King: P (A) = 52
= 13
.
P (A∩B) 4/52
(b) To draw a King knowing that face card has been drawn: P (A|B) = P (B)
= 12/52
= 13 .

Problem 8. Men and women often disagree on how they think about selecting a mate. Suppose that
a poll of 1000 individuals in their twenties gave the following responses to the question of whether it
is more important for their future mate to be able to communicate their feelings (F ) than it is for that
person to make a good living (G).

Feelings (F) Good Living (G) Totals

Men (M) .35 .20 .55
Women (W) .36 .09 .45
Totals .71 .29 1.00

If an individual is selected at random from this group of 1000 individuals, calculate the following prob-
abilities:

(a) P (F ) (b) P (G) (c) P (F |M ) (d) P (F |W ) (e) P (M |F ) (f )P (W |G) .

Solution:
.35
(a) P (F ) = .71 (b) P (G) = .29 (c) P (F |M ) = = 0.636 ,
.55
.36 .35 .09
(d) P (F |W ) = = 0.8 (e) P (M |F ) = = 0.493 (f )P (W |G) = = 0.31 .
.45 .71 .29

Problem 9. A certain manufactured item is visually inspected by two different inspectors. When a
defective item comes through the line, the probability that it gets by the first inspector is 0.1. Of those
that get past the first inspector, the second inspector will ”miss” 5 out of 10. What fraction of the
defective items get by both inspectors?

6
Solution: Let A be the event a defective item gets by the first inspector, and B be the event a
defective item gets by the second inspector. We are given P (A) = 0.1 and P (B|A) = 0.5. The
question is about P (A ∩ B). It is found as
P (A ∩ B) = P (B|A) × P (A) = 0.5 × 0.1 = 0.05 .

Recall that two events A and B are independent if the fact that A occurs does not affect the
probability of B occurring. Formally this means that P (B) = P (B|A). It is easy to that if
P (A) ̸= 0 ̸= P (B), then A and B are independent if and only if
P (A ∩ B) = P (A)P (B) .
Problem 10. On a typical February day in Sofia the probability of probability of snowing is 0.1, the
probability of encountering a traffic jam is .8, and the probability of snowing or encountering a traffic
jam (or both) is .82. Are the events ”snowing” and the event ”encountering a traffic jam” independent?
Solution: Let A be the event “snowing”, and B be the event “traffic jam”. We are given that
P (A) = 0.1, P (B) = 0.8 and P (A ∪ B) = 0.82. From here
P (A ∩ B) = P (A) + P (B) − P (A ∪ B) = 0.8 + 0.1 − 0.82 = 0.08 .
Since P (A) × P (B) = 0.8 × 0.1 = 0.08 = P (A ∩ B), it follows that the events “snowing” and
“traffic jam” are independent.
Problem 11. Suppose that P (A ∪ B) = 0.9, P (A) = 0.5 and P (B) = 0.6. Are A and B independent?
Solution: From P (A ∪ B) = P (A) + P (B) − P (A ∩ B) we find that
P (A ∩ B) = P (A) + P (B) − P (A ∪ B) = 0.5 + 0.6 − 0.9 = 0.2 .
Since P (A)P (B) = 0.5 × 0.6 = 0.3 ̸= 0.2 = P (A ∩ B), we conclude that A and B are not
independent, i.e., they are dependent events.
Problem 12. Suppose that P (A ∪ B) = 0.8, P (A) = 0.5 and P (B) = 0.6. Are A and B independent?
Solution: From P (A ∪ B) = P (A) + P (B) − P (A ∩ B) we find that
P (A ∩ B) = P (A) + P (B) − P (A ∪ B) = 0.5 + 0.6 − 0.8 = 0.3 .
Since P (A)P (B) = 0.5 × 0.6 = 0.3 = P (A ∩ B), we conclude that A and B are independent.

1.4 Counting rules

1.4.1 The mn-rule

Let A = {x1 , x2 , . . . , xm } be a set of m elements and let B = {y1 , y2 , . . . , yn } be a set of n elements.

Then there are mn ordered pairs,
(xi , yj ), i = 1, 2, . . . , m and j = 1, 2, . . . , n,
of elements, the first from A and the second from B. Informally, we often speak of ways to
select such a pair.
For this reason it is called called the mn−rule (m times n-rule), or also the multiplication rule.
Example 13. There are four blood types, A, B, AB, and O. Blood sample can also be Rh factors, i.e.
Rh+ and Rh−. Finally, a blood donor can be classified as either male or female. How many different
ways can a donor have his or her blood labeled?

7
 Figure 1: The tree diagram for the outcomes when blood type and Rh factor are recorded.

1.4.2 The extended mn-rule

If Aj , j = 1, . . . , n is a set with mj elements, say Aj = {xj,1 , xj,2 , . . . , xj,mj }, then there are
m1 × m2 × · · · × mn ordered n−uples,

(x1,i1 , x2,i2 , . . . , xn,in ), 1 ≤ i1 ≤ m1 , . . . , 1 ≤ in ≤ mn .

If in the previous example we account also the gender of the experimental unit, then the num-
ber of different recordings that can be register will double.

Figure 2: The tree diagram when blood type, Rh factor and gender are recorded.

Example 14. You have three groups of distinctly different items, four in the first group, seven in the
second, and three in the third. If you select one item from each group, how many different triplets can
you form?

Solution: Answer: 4 × 7 × 3 = 84.

8
Example 15. A particular cell phone company offers 4 models of phones, each in 6 different colors and
each available with any one of 5 calling plans. How many combinations are possible?

Solution: Answer: 4 × 6 × 5 = 120.

1.4.3 Permutations

The factorial notation uses the exclamation point:

n! = n × (n − 1) × · · · × 2 × 1 .

Also, by definition 0! = 1.

Definition 16. A permutation is an arrangement of n objects in a specific order.

Definition 17. The arrangement of n objects in a specific order using r objects at a time is called a
permutation of n objects taking r objects at a time. It is written as nP r, or Prn , and the formula is

n!
nP r = Prn = .
(n − r)!

Example 18. Books on a bookshelf:

Figure 3: Arranging books

• How many are the books? A: 25

• In how many different ways can we arrange them? A:

25 · 24 · · · 2 · 1 = 25! = 1.5511 × 1025

• In how many different ways can we arrange precisely 6 of them? A:

25!
25 · 24 · 23 · 22 · 21 · 20 = = P625 = 127512000
19!
Example 19. A radio talk show host can select 4 of 6 special guests for her program. The order of
appearance of the guests is important. In how many different ways can this be done?
6 6!
Answer: P6−4 = 2!
= 360 .

9
1.4.4 Permutations of non-distinct objects

Suppose that we are given a set of n objects some of which are indistinguishable. It is not
difficult to see that the number of permutations of n objects when r1 objects are identical, r2
objects are identical, . . . , rk objects are identical, etc., is

n!
r1 ! × r2 ! × · · · × rk !
where r1 + r2 + · · · + rk = n.

Example 20. How many permutations of the letters can be made from the word STATISTICS?

10!
Answer: = 50400.
3! × 3! × 2! × 1! × 1!

1.4.5 Combinations

A selection of distinct objects without regard to order is called a combination.

Example 21. Given 25 books. In how many ways can we select 6 ignoring the order?

Solution: If we account order:

25!
25 · 24 · 23 · 22 · 21 · 20 = = P625 = 127512000
19!
as every selection would have been counted 6! times. Therefore if we ignore the order we must
divide by 6!, so the number of different ways is

25 · 24 · 23 · 22 · 21 · 20 25! 6
= = C25 = 177100
6·5·4·3·2·1 19! · 6!

The number of combinations of r objects selected from n objects is denoted by nCr, or Crn , or
Cnr , or nr and is given by the formula

 
r n n!
nCr = Cn = = .
r (n − r)!r!

Example 22. How many combinations of 6 objects are there, taken 2 at a time?
6


Answer: 2
= 15.

Example 23. In a club there are 7 women and 5 men. A committee of 3 women and 2 men is to be
chosen. How many different possibilities are there?
7 5



Answer: 3
× 2
= 35 × 10 = 350.

Example 24. Five cards are selected from a 52-card deck for a poker hand.

(a) How many simple events are in the sample space?

10
 (b) A royal flush is a hand that contains the A, K, Q, J, and 10, all in the same suit. How many ways
are there to get a royal flush?

(c) What is the probability of being dealt a royal flush?

Solution:
52


(a) 5
= 2598960

(b) 4
4
(c) 2598960
= 0.0000015391

Problem 25 (Coffee Selection). A coffee shop serves 12 different kinds of coffee drinks. How many
ways can 4 different coffee drinks be selected?
12


Solution: Answer: 4
= 495.

Example 26. Five cards are selected from a 52-card deck for a poker hand. You have a poker hand
containing four of a kind.

(a) How many possible poker hands can be dealt?

(b) In how many ways can you receive four cards of the same face value and one card from the other
48 available cards?

(c) What is the probability of being dealt four of a kind?

Solution:
52


(a) 5
= 2598960

(b) 13 × 48
13×48
(c) 2598960
= 0.00024010

Example 27. Find the probability of getting 4 aces when 5 cards are drawn from an ordinary deck of
cards.

Solution:
4 48



4
× 1 48
P (4 aces) = 52

= = 0.0000184689 .
5
2598960

11
1.5 Summary of the counting rules and more problems
Let A be a set with n elements.

• there are n × n × · · · × n = nk such k−tuples whose components are elements of A;

• to compute the number of k−tuples whose components are distinct elements of A, we get

n!
n × (n − 1) × (n − 2) × · · · × (n − (k − 1)) = .
(n − k)!
n!
We call each such k−tuple a permutation and denote it by Pkn , i.e. Pkn = (n−k)!
.

• If the order is not important, we just count the number of subsets of k elements taken
from A. We obtain it as
n × (n − 1) × (n − 2) × · · · × (n − (k − 1)) Pn n!
= k = .
k! k! (n − k)!k!

We denote it by nk and say it gives the number of combinations of k elements taken from a

set of n elements.

Example 28 (Hawaiian Words). The Hawaiian alphabet consists of 7 consonants and 5 vowels. How
many three-letter “words” are possible if there are never two consonants together and if a word must
always end in a vowel?

Solution:

7 × 5 × 5 + 5 × 12 × 5 = 4975 .

Problem 29. What is the probability that at least two students in STA 1005 have the same birthday?
Assume that the number of students is n = 30.

Solution: Denote by A the event there are at least two students born on the same day. We will
calculate P (Ac ). In fact, Ac is the event there are NO students born on the same day. It is easy to
365! 365
find that nAc = 365 × 364 × · · · × 336 = = P30 , while the total number of possible
(365 − 30)!
outcomes is N = 36530 . Therefore,

30! 365


c Pn365 30
P (A) = 1 − P (A ) = 1 − =1− ≈ 70.6% .
365n 36530

Problem 30. What is the probability that exactly two students in STA 1005 have the same birth-
day?1 Assume that the number of students is n = 30.
1
Notice the difference here!

12
Solution: Denote by A the event there are exactly two students born on the same day. Then it is
easy to calculate that
nA = C2n × 365 × Pn−2364
,
the explanation being that

• we can select two students out of n in C2n different ways;

• we can select a birthday for them in 365 different ways;

• the remaining n − 2 students should be all born on different days, and as there are 364
364
remaining available days, there are Pn−2 different ways for this to be done.

Therefore,
30 × 29 364!
364
C n × 365 × Pn−2 × 365 ×
P (A) = 2 = 2 336! .
365n 36530

Problem 31. A hand of 5 cards is selected (the order is ignored) from an ordinary deck of 52 cards.

(a) How many are the hands containing two aces, two black kings, and a jack?

(b) How many are the hands such that exactly three of the cards are diamonds?

(c) How many are the hands containing 5 cards of three different values, i.e. either one of the values is
repeated 3 times and the others are different, or two different are repeated twice and another value
appears once?

Solution:

(a) The number of different hands containing two aces, two black kings, and a jack is
     
4 2 4
× × = 24 .
2 2 1

(b) The number of hands such that exactly three of the cards are diamonds is
   
13 39
× = 211926 .
3 2

(c) To count correctly the number of hands containing 5 cards of three different values, we
count separately:

N2,2,1 the number of hands in which there are two values repeated twice, and a 5-th card
of different value,
N3,1,1 the number of hands in which there are one values repeated three times, and a the
4-th and 5-th cards being of different values.

13
          
13 4 4 11 4
N2,2,1 = × × × × = 123552
2 2 2 1 1
         
13 4 12 4 4
N3,1,1 = × × × × = 54912,
1 3 2 1 1
hence

the

number of
hands

containing
5 cards

of three different values is N2,2,1 + N3,1,1 =
13 4 4 11 4 13 4 12 4 4





2
× 2 × 2 × 1 × 1 + 1 × 3 × 2 × 1 × 1 = 123552 + 54912 = 178464.

2 Additional problems on probability

Problem 32. Suppose P (A) = 0.1 and P (B) = 0.5.

(a) If P (A|B) = 0.1, what is P (A ∩ B)?

(b) If P (A|B) = 0.1, are A and B independent?
(c) If P (A ∩ B) = 0, are A and B independent?
(d) If P (A ∪ B) = 0.65, are A and B mutually exclusive?

Solution: Here are the answers:

(a) Since P (A ∩ B) = P (A|B) × P (B) = 0.1, we find P (A ∩ B) = 0.1 × 0.5 = 0.05.

(b) Since P (A|B) = 0.1 = P (A) it follows that A and B are independent.
(c) Since P (A ∩ B) = 0 ̸= 0.05 = P (A) × P (B), the events A and B couldn’t possibly be
independent.
(d) If A and B were mutually exclusive then we would have P (A ∩ B) = 0, hence P (A ∪ B) =
P (A) + P (B) − P (A ∩ B) = P (A) + P (B) = 0.6. Since we are given that P (A ∪ B) = 0.65,
it follows that the events A and B are not mutually exclusive.

Problem 33. Player A has entered a golf tournament but it is not certain whether player B will enter.
Player A has probability 1/6 of winning the tournament if player B enters and probability 3/4 of win-
ning if player B does not enter the tournament. If the probability that player B enters is 1/3, find the
probability that player A wins the tournament.

Solution: Let A be the event player A wins the tournament, and B be the event player B enters
participates in the tournament. We are given P (A|B) = 1/6, P (A|B c ) = 3/4 and P (B) = 1/3
(hence P (B c ) = 2/3). The question is about P (A). We find this probability as follows

P (A) = P (A ∩ B) + P (A ∩ B c ) = P (A|B) × P (B) + P (A|B c ) × P (B c ) =

1 1 3 2 5
= × + × = .
6 3 4 3 9

14
3 Probability Distributions
Example 34. A box contains 24 integrated circuits, 4 of which are defective. If 4 are sold at random,
find the following probabilities.

(a) Exactly 2 are defective.

(b) None is defective.

(c) All are defective.

(d) At least 1 is defective.

24


Solution: Denote by n = number ways to select 4 out of 24. Easy to see that n = 4
= 10626.

20 4



(a) To get exactly 2 defective: 2
× 2
= 190 × 6 = 1140, so
1140
P ( exactly 2 defective ) = = 0.10728
10626
20 4



(b) To get no defectives: 4
× 0
= 4845 × 1 = 4845, so
4845
P ( no defectives ) = = 0.45596
10626
20 4



(c) To get only defective: 0
× 4
= 1 × 1 = 1, so
1
P ( only defective ) = = 0.000094109
10626

(d) To compute the probability for at least 1 defective: denote B : the event “no defectives”,
so want P (B̄) = 1 − P (B) = 1 − 0.45596 = 0.54404.

Definition 35. A variable X is a random variable if the value that it assumes, corresponding to the
outcome of an experiment, is a chance or random event.

For example:

1. X = Number of times a student talks on the phone on a given day

2. X = SAT score for a randomly selected college applicant

In general we want to know:

• What values of X occur?

• How often each value of X occurs?

Recall that a variable is called:

• discrete if it can assume only finitely many values, or countably many infinite values

15
 • continuous if it can assume any value in an interval between between two different real
numbers

A discrete distribution describes the probability of occurrence of each value of a discrete random
variable. A discrete random variable is a random variable that takes values in a set with finitely
many or countably many elements, such as a list of non-negative integers.

Definition 36. A discrete probability distribution consists of the values a random variable can -
assume and the corresponding probabilities of the values.

The probabilities are determined theoretically or by observation.

How to find a probability distribution?

1. Make a frequency distribution for the outcomes of the variable.

2. Find the probability for each outcome by dividing the frequency of the outcome by the
sum of the frequencies.

3. If a graph is required, place the outcomes on the x-axis and the probabilities on the y axis,
and draw vertical bars for each outcome and its corresponding probability.

Example 37. If we roll a fair die once then

X=k 1 2 3 4 5 6
1 1 1 1 1 1
P (X = k) 6 6 6 6 6 6

Example 38. If we trow 3 identical coins and X counts the number of H’s (heads):

X=k 0 1 2 3
1 3 3 1
P (X = k) 8 8 8 8

1. The probability of each event in the sample space must be between or equal to 0 and 1.
That is, 0 ≤ P (X) ≤ 1.

2. P
The sum of the probabilities of all the events in the sample space must equal 1, i.e.
P (X) = 1.

Example 39. Determine whether the given function P (X) could possibly be a probability distribution.

16
 X=k ♣ ♢ ♡ ♠
(a)
1 1 1 1
P (X = k) 4 2 8 8

(b) The probabilities that a student is registered for X = 2, 3, 4 or 5 are 0.01, 0.34, 0.62, and 0.03,
respectively.

X=k 2 5 9 17
(c)
P (X = k) 0.5 0.35 −0.2 0.35

Problem 40. Past experience has shown that, on the average, only 1 in 10 wells drilled hits oil. Let
x be the number of drillings until the first success (oil is struck). Assume that the drillings represent
independent events.

(a) Find p(1), p(2), and p(3).

(b) Give a formula for p(x).

(c) Graph p(x).

Solution: We give just short answers, the reasoning being obvious.

(a) p(1) = 1/10 = 0.1, p(2) = 9/10 × 1/10 = 0.09, and p(3) = 9/10 × 9/10 × 1/10 = 0.081.
 k−1
9 1
(b) Easily, p(x = k) = × .
10 10
(c) Here is the graph for p(x).

Figure 4: Probability distribution of finding oil at the k-th try.

Problem 41. A random variable x can assume five values: 0, 1, 2, 3, 4. A portion of the probability
distribution is shown here:

x 0 1 2 3 4
p(x) .1 .3 .3 ? .1

17
 (a) Find p(3).
(d) What is the probability that x is greater than 2?
(e) What is the probability that x is 3 or less?

Solution:

(a) Since the sum of the probabilities must equal 1, we have p(3) = 0.2.
(d) The probability that x is greater than 2 is equal to p(3) + p(4) = 0.3.
(e) The probability that x is 3 or less is the sum p(0)+p(1)+p(2)+p(3) = 0.1+0.3+0.3+0.2 =
0.9.

3.1 Mean, variance, standard deviation, and expectation

Example 42. If you roll a single die then

x=k 1 2 3 4 5 6
1 1 1 1 1 1
P (x) = k 6 6 6 6 6 6

1 1
µ=1× + · · · + 6 × = 3.5 .
6 6

1 1 1 1
σ 2 = (3.5 − 1)2 × + (3.5 − 2)2 × + · · · + (3.5 − 5)2 × + 6 × = 2.91667 .
6 6 6 6
√
Hence σ = σ 2 = 1.70783.

Example 43. A talk radio station has four telephone lines. If the host is unable to talk (i.e., during a
commercial) or is talking to a person, the other callers are placed on hold. When all lines are in use, others
who are trying to call in get a busy signal. The probability that 0, 1, 2, 3, or 4 people will get through is
shown in the probability distribution. Find the variance and standard deviation for the distribution.

x 0 1 2 3 4
p(x) 0.18 0.34 0.23 0.21 0.04

Solution:
P
• Mean: µ = xp(x) = 1.59;
• Variance: µ = (x − µ)2 p(x) = x2 p(x) − ( xp(x))2 = 3.79 − 1.592 = 1.2619;
P P P
√
• Standard deviation: σ = σ 2 = 1.1233

18
3.2 Computing expected values
Problem 44. You can insure a $50,000 diamond for its total value by paying a premium of D dollars.
If the probability of loss in a given year is estimated to be 0.01, what premium should the insurance
company charge if it wants the expected gain to equal $1000?

Solution: Denote by X the amount of money the insurance company pays for every insured set
of diamonds. In the given setup, X can take only two values: X = 0 – in case of no insurance
event, and X = 50, 000 – in case of loss of the diamonds. We are given P (X = 0) = 0.99 and
P (X = 50, 000) = 0.01. Hence the expected payback for a randomly chosen set of diamonds is

E(X) = 0 × 0.99 + 50, 000 × 0.01 = 500 ,

therefore if the insurance company wants to gain 1000 on average for each set of diamonds, it
needs to charge a premium of 500 + 1, 000 = $1, 500.

Example 45. Members of the student government are planning a movie night on campus. They have
asked a random sample of students which of four movies the students would prefer. Here are the results.

Movie Preferred Number of Students

Movie A 63
Movie B 20
Movie C 36
Movie D 21

There are 800 students total on campus. Based on the above information, how many students on campus
would we expect to prefer Movie A? Round your answer to the nearest whole number. Do not round
any intermediate calculations.
Hypothetically, there might be students who are not interested in any of these movies, which is not
accounted in the sample. Thus, to answer the question, we will assume that each of the 800 students will
favour one of those movies. Then, the expected number of students on campus who would be expected to
prefer Movie A is
63
E(XA ) = × 800 = 0.45 × 800 = 360 .
63 + 20 + 36 + 21

Example 46. Digitalis is a technology company that makes high-end computer processors. Their newest
processor, the luteA, is going to be sold directly to the public. The processor is to be sold for $2700,
making Digitalis a profit of $464. Unfortunately there was a manufacturing flaw, and some of these
luteA processors are defective and cannot be repaired. On these defective processors, Digitalis is going
to give the customer a full refund. Suppose that for each luteA there is a 13% chance that it is defective
and an 87% chance that it is not defective.
If Digitalis knows it will sell many of these processors, should it expect to make or lose money from
selling them? How much?
To answer the question, we take into account the profit earned on each processor and the expected value
of the amount refunded due to the processor being defective. Denote by X the profit from each processor.
Its expected value is

19
 E(X) = 0.87 × 464 − 0.13 × (2700 − 464) = 113.
Therefore, Digitalis can expect to make money from selling these processors. In the long run they should
expect to make $113 dollars on each processor sold.

Example 47. Aldo is playing a game in which he spins a spinner with 6 equal-sized slices numbered 1
through 6. The spinner stops on a numbered slice at random. This game is this: Aldo spins the spinner
once. He wins $1 if the spinner stops on the number 1, $3 if the spinner stops on the number 2, $5 if the
spinner stops on the number 3, and $7 if the spinner stops on the number 4. He loses $8 if the spinner
stops on 5 or 6. What is the expected value (gain/loss) of playing the game?

To find the expected value of playing the game, we use that probability that spinner stops on each num-
bered slice is same.
1 1 1 1 1 1
E(X) = 1 × + 3 × + 5 × + 7 × − 8 × − 8 × = 0.
6 6 6 6 6 6
Thus, Aldo can expect to break even (neither gain nor lose money).

20
4 Several discrete probability distributions
4.1 The Binomial distribution
A binomial experiment consists of n identical trials with probability of success p on each trial.
The probability of k successes in n trials is

n!
P (X = k) = Ckn pk q n−k = pk q n−k
k!(n − k)!

for values of k = 0, 1, 2, ..., n.

Mean: µ = np
Variance: σ 2 = npq
√
Standard deviation: σ = npq

Problem 48. Car color preferences change over the years and according to the particular model that the
customer selects. In a recent year, suppose that 10% of all luxury cars sold were black. If 25 cars of that
year and type are randomly selected, find the following probabilities:

(a) At least five cars are black.

(b) At most six cars are black.

(c) More than four cars are black.

(d) Exactly four cars are black.

(e) Between three and five cars (inclusive) are black.

(f) More than 20 cars are not black.

Solution: Let X be the number of black cars in the sample of 25. Then we have a binomial
experiment with probability for a success in a single trial p = 0.1 and X counts the number of
successes. In terms of probabilities about X the questions are:

(a) P (X ≥ 5) = 1 − P (X ≤ 4) = 1 − 0.902 = 0.098.

(b) P (X ≤ 6) = 0.991.

(c) P (X > 4) = P (X ≥ 5) = 1 − P (X ≤ 4) = 1 − 0.902 = 0.098.

(d) P (X = 4) = P (X ≤ 4) − P (X ≤ 3) = 0.902 − 0.764 = 0.138.

(e) P (3 ≤ X ≤ 5) = P (X ≤ 5) − P (X ≤ 2) = 0.967 − 0.537 = 0.43.

(f) P (25 − X > 20) = P (X < 5) = P (X ≤ 4) = 0.902.

Problem 49. A home security system is designed to have a 99% reliability rate. Suppose that nine homes
equipped with this system experience an attempted burglary. Find the probabilities of these events:

21
(a) At least one of the alarms is triggered.
(b) More than seven of the alarms are triggered.
(c) Eight or fewer alarms are triggered.

Solution: Here we have a binomial experiment with n = 9 with probability for success p =
0.99. Denote by X the number of successes, that is, the number of houses in which the alarm is
triggered.

(a) At least one of the alarms is triggered is equivalent to ask for P (X ≥ 1).

P (X ≥ 1) = 1 − P (X ≤ 0) ≈ 1 − 0 = 1 .

(b) More than seven of the alarms to be triggered is equivalent to ask for P (X > 7), so

P (X > 7) = 1 − P (X ≤ 7) = 1 − 0.003 = 0.997 .

(c) Eight or fewer alarms to be triggered is equivalent to ask for P (X ≤ 8).

P (X ≤ 8) = 0.086 .

Problem 50. Forty percent of all Americans who travel by car look for gas stations and food outlets that
are close to or visible from the highway. Suppose a random sample of n = 25 Americans who travel by
car are asked how they determine where to stop for food and gas. Let X be the number in the sample who
respond that they look for gas stations and food outlets that are close to or visible from the highway.

(a) What are the mean and variance of X?

(b) Calculate the interval µ±2σ. What values of the binomial random variable X fall into this interval?
(c) Find P (6 ≤ X ≤ 14). How does this compare with the fraction in the interval µ ± 2σ for any
distribution? For mound-shaped distributions?

Solution: Given is p = 0.4, hence q = 0.6. Also, n = 25.

(a) The mean is µ = µX = n × p = 10 and the variance is σ 2 = σX 2

= n × p × q = 6.
√
(b) We have that σ = σ 2 = 2.44949. The interval µ ± 2σ is µ ± 2σ ≡ [µ − 2σ, µ + 2σ] ≡
[5.101, 14.899]. The second question: What values of the binomial random variable X fall into
this interval?, should be understood as, What is P (5.101 ≤ X ≤ 14.899) ?. Thus, the answer
is: P (5.101 ≤ X ≤ 14.899) = P (X ≤ 14) − P (X ≤ 5) = 0.966 − 0.029 = 0.937 ≈ 94%.
(c) The probability P (6 ≤ x ≤ 14) was in fact already calculated in (b), i.e. P (6 ≤ X ≤
14) = P (X ≤ 14) − P (X ≤ 5) = 0.937. This means that in this case (X−binomial with
n = 25 and p = .4), P (µ − 2σ ≤ X ≤ µ + 2σ) = P (6 ≤ X ≤ 14) = 0.937 ≈ 94%. If
we to compare this result with the one about a random variable, say X̃, with mound-
shaped distribution we are to observe that P (µX − 2σX ≤ X ≤ µX + 2σX ) ≈ 94% and
P (µX̃ −2σX̃ ≤ X̃ ≤ µX̃ +2σX̃ ) ≈ 95%, which are close enough for some practical purposes.
So, this suggests that when n is large, we could approximate the binomial distribution with the
mound-shaped distribution.

22
Problem 51. A psychiatrist believes that 80% of all people who visit doctors have problems of a psycho-
somatic nature. She decides to select 25 patients at random to test her theory.

(a) Assuming that the psychiatrist’s theory is true, what is the expected value of X, the number of the
25 patients who have psychosomatic problems?

(b) What is the variance of X, assuming that the theory is true?

(c) Find P (X ≤ 14). (Use tables and assume that the theory is true.)

(d) Based on the probability in part (c), if only 14 of the 25 sampled had psychosomatic problems, what
conclusions would you make about the psychiatrist’s theory? Explain.

Solution: As X is defined as the number of patients who have psychosomatic problems in the
sample of n = 25, it must have binomial distribution with probability for success p = 0.8, and
for failure q = 0.2.

(a) The expected value of X is E(X) = µx = µ = np = 25 × 0.8 = 20, that is, it is expected that
20 out of the sample of 25 will have psychosomatic problems.

(b) The variance of X is E(X − µ)2 = npq = 25 × 0.8 × 0.2 = 4.

(c) P (X ≤ 14) = 0.006.

(d) Suppose that it turns out that X = 14; since the probability for this is P (X = 14) = P (x ≤
14) − P (x ≤ 13) = 0.006 − 0.002 = 0.004, i.e., it is rather small number, we might actu-
ally question whether the psychiatrist’s hypothesis that 80% of all patients have psychotic
problems. That is, one can question if p = 0.8 is really the true value.

4.2 The Poisson Distribution

In probability and statistics, the Poisson distribution was introduced by the French mathemati-
cian Siméon Denis Poisson. It is a discrete probability distribution that expresses the proba-
bility of a given number of events occurring in a fixed interval of time or space provided that
these events occur with at a (known) constant mean rate and independently of one another.
The Poisson distribution can also be used for the number of events in other specified intervals
such as distance, area or volume.
Summary of the requirements:

1. The random variable X is the number of occurrences of an event over some interval (i.e.,
length, area, volume, period of time, etc.).

2. The occurrences occur randomly.

3. The occurrences are independent of one another.

4. The average number of occurrences over an interval is known.

23
The probability of k occurrences in an interval of time / volume / area, etc., counted by a
variable X, given that the mean (average) number of occurrences per unit (time, volume, area,
etc.) is λ (Greek letter lambda), is given by

e−λ λk
P (X = k) =
k!
where k = 0, 1, . . ..
Remark: In a Poisson experiment

• X is the number of events that occur in a period of time or space during which an average
of λ such events can be expected to occur.

• The only assumption is that the events occur randomly and independently of one another.
e−λ λk
• P (X = k) = k!
for k = 0, 1, . . ..
∞
X e−λ λk
• E(X) = µX = k = λ, that is, E(X) = µX = µ = λ .
k=0
k!
∞
X e−λ λk
• E(X − E(X))2 = σX
2
= (k − λ)2 = λ, that is,
k=0
k!
√
E(X − E(X))2 = σX
2
=λ and σX = λ.

Remark: In many texts instead of λ the average number of occurrences is denoted by µ.

Problem 52. The number X of people entering the intensive care unit at a particular hospital on any
one day has a Poisson probability distribution with mean equal to four people per day.

(a) What is the probability that the number of people entering the intensive care unit on a particular
day is two?

(b) What is the probability that the number of people entering the intensive care unit on a particular
day is at most two?

(c) What is the probability that the number of people entering the intensive care unit will exceed 10
over a period of two days?

Solution:

(a) The probability that the number of people entering the intensive care unit on an arbitrary
chosen day to be two is

µ2X e−µX 42 e−4

P (X = 2) = = = 0.1465251 .
2! 2

(b) The probability that the number of people entering the intensive care unit on an arbitrary
chosen day to be at most two is obtained as

P (X ≤ 2) = 0.238 .

24
 (c) To calculate the probability that the number of people entering the intensive care unit will
exceed 10 over a period of two days, we denote by Y the number of people entering the
intensive care unit over a period of two days. Then Y will have Poisson distribution with
µY = 4 + 4 = 8. The the probability that the number of people entering the intensive care
unit will exceed 10 over a period of two days is

P (Y > 10) = 1 − P (Y ≤ 10) = 1 − 0.816 = 0.184 .

Example 53. According to a study conducted by the Department of Pediatrics at the University of
California, San Francisco, children who are injured two or more times tend to sustain these injuries
during a relatively limited time, usually 1 year or less. If the average number of injuries per year for
school-age children is two, what are the probabilities of these events?

(a) A school-age child will sustain two injuries during the year.

(b) A school-age child will sustain two or more injuries during the year.

(c) A school-age child will sustain at most one injury during the year.

Solution: Here we have an example of a random variable, X measuring the number of injuries
per year for a school-age child. Such X has Poisson distribution with λ = µX = 2. The
questions/ answers are:

(a) P (X = 2) = P (X ≤ 2) − P (X ≤ 1) = 0.677 − 0.406 = 0.271.

(b) P (X ≥ 2) = 1 − P (X ≤ 1) = 1 − 0.406 = 0.594.

(c) P (X ≤ 2) = 0.677.

4.3 Approximating Binomial distribution with Poisson distribution

Suppose we are presented with a binomial experiment with n trials in which:

• P (S) = p, P (F ) = q = 1 − p

• n is “LARGE”

• p is “small”

• np = λ is neither “LARGE” nor “small”, i.e. 0 < λ = np < 7.

Denote, as usual, X = number of successes in the binomial experiment and consider the task
of finding P (X = k) = nk pk q n−k , where k is “small” and n is “LARGE”. As we showed at

lectures, the probability P (X = k) is approximated well by the probability P (Y = k), where Y

is a random variable with Poisson distribution with λ = µY = np. Summarizing, if X counts
the number of successes in a binomial experiment with n trials and probability for success p
for which

25
 • n ≥ 30 and
• np < 7

then
P (X = k) ≈ P (Y = k) and P (X ≤ k) ≈ P (Y ≤ k) ,
where Y is a random variable counting the number of occurrences in a Poisson experience for
which µY = λ = np.
Problem 54. An insurance company is considering the addition of major-medical coverage for a rela-
tively rare ailment. The probability that a randomly selected individual will have the ailment is 0.001,
and 3 000 individuals are included in the group that is insured.

(a) What is the expected number of people who will have the ailment in the group?
(b) What is the probability that no one in this group of 3 000 people will have this ailment?
(c) What is the probability that between two and five individuals (inclusive) in this group of 3 000
people will have this ailment?

Solution: The given n = 3 000 individuals can be considered as experimental units in a bi-
nomial experiment in which a person is checked if he / she has that specific ailment. The
probability for “success”, i.e. for a person to have the ailment, is given to be p = 0.001 and
thus the probability for “failure” is q = 1 − p = 0.999. As usual, denote by X the number of
successes in the binomial experiment, i.e. the number of people who have the ailment.

(a) The expected number of people who will have the ailment in the group is

µ ≡ µX = np = 3 .

(b) The probability that no one in this group of 3 000 people will have this ailment is given
by
P (X ≤ 0) = P (X = 0) .
In general, probabilities of this kind should be approximated. Since n = 3000 > 30 and
np = 3 < 7
P (X ≤ 0) ≈ P (Y ≤ 0) ,
where Y is a random variable with Poisson distribution for which µY = np = 3. Thus, we
find using [Blu2017, Table C in Appendix A], or computing directly with Excel, that

P (X = 0) ≈ P (Y = 0) = 0.49787068 .

(c) The probability that between two and five individuals (inclusive) in this group of 3 000
people will have this ailment is is given by

P (2 ≤ X ≤ 5) = P (X ≤ 5) − P (X ≤ 1) .

Since n = 3000 > 30 and np = 3 < 7, as explained in (b), probabilities of the kind
P (X ≤ k), are approximated with P (Y ≤ k), where Y is the same random variable with
Poisson distribution as in (b). Therefore, using again Excel, or the ALEKS calculator, or
[Blu2017, Table C in Appendix A], we find

P (2 ≤ X ≤ 5) ≈ P (Y ≤ 5) − P (Y ≤ 1) = 0.916082 − 0.199148 = 0.716934 .

26
Problem 55. Evidence shows that the probability that a driver will be involved in a serious automobile
accident during a given year is 0.01. A particular corporation employs 100 full-time traveling sales reps.
Based on this evidence, use the Poisson approximation to the binomial distribution to find the probability
that exactly two of the sales reps will be involved in a serious automobile accident during the coming
year

Solution: Let X be number of sales reps involved in a serious automobile accident during a
year. So, X has binomial distribution with p = 0.01. We have to find P (X = 2). Since the
sample size is n = 100 and np = 1 < 7, we can approximate P (X = 2) using a random variable
Y with Poisson distribution for which λ = µY = np = 1. Therefore, P (X = 2) ≈ P (Y = 2) =
0.1839.

27
5 Continuous distributions
5.1 Basics on continuous distributions
Consider X : continuous random variable.

• The cumulative distribution function F (x0 ) for a random variable X is by definition the prob-
ability
F (x0 ) = P (X ≤ x0 )
where −∞ < x0 < ∞.

• The probability density function for a random variable X is a theoretical model for this distri-
bution that, roughly said, gives the rate of change of F (x) at x. More precisely, if F (x) is
the cumulative distribution function for a continuous random variable X, then the density
function f (x) for X is
dF (x)
f (x) = .
dx
Example 56. The probability density function of the continuous uniform distribution is:

1
 b−a for a ≤ x ≤ b,
f (x) =
0 for x < a or x > b
The notation used for such an r.v. X is X ∼ U (a, b).

Example 57. A real number between 1 and 11 will be picked according to the probability distribution
shown in the figure. Regions under the curve are labeled with A, B, and C.

Figure 5: The regions A, B, and C.

The area of each region is as follows: area(A) = 0.2, area(B) = 0.6, and area(C) = 0.2.

(a) The probability that a real number between 1 and 3 is picked is 0.2.

(b) The probability that a real number between 1 and 3 is picked is 0.8.

28
 Figure 6: The graph Γf of the probability density function fX (x).

Example 58. The graph of a probability distribution is shown below. The graph is made up of two line
segments and a curved path.
For each part below, suppose a real number is picked according to the above probability distribution.

(a) Find the probability that the number picked is between 6 and 6.

(b) Find the probability that the number picked is between 4.3 and 5.7.

(c) Find the probability that the number picked is 4.

Denote by X the random variable that gives the real number that is selected. What we are shown in blue
is the graph Γf of the probability density function f (x) := fX (x). The probability P (a ≤ X ≤ b) is the
area under Γf on the interval [a, b]. Therefore, to answer the above questions we need to calculate the
corresponding areas.
“Reading” the graph and using the elementary formulae for areas, we find that:

(a) the probability that the number picked is between 6 and 10 is P (6 ≤ X ≤ 10) = 1 − P (1 ≤ X ≤
6) = 1 − (P (1 ≤ X ≤ 3) + P (3 ≤ X ≤ 6)) = 1 − ( (3−1)×0.15
2
+ (6 − 3) × 0.15) = 1 − 0.6 = 0.4.

(b) The probability that the number picked is between 4.3 and 5.7 is P (4.3 ≤ X ≤ 5.7) = (5.7 − 4.3) ×
0.15 = 0.21.

(c) The probability that the number picked is 4 is P (X = 4) = 0.

29
5.2 The Normal distribution
A continuous random variable X is said to have normal distribution with mean µ and standard
deviation σ, notation X ∼ N (µ, σ 2 ), if
1 (x−µ)2
• its probability density function is f (x) = √ e− 2σ2 , and
σ 2π
Rx0 1 (x−µ)2
• the cumulative distribution function F (x0 ) for X is F (x0 ) = P (X ≤ x0 ) = √ e− 2σ2 dx
−∞ σ 2π
• Visually

f (x)

µ x0 x

The normal distribution is known also as Gauss distribution:

Figure 7: Gauss and the curve of errors on a 10 Deutschmark note

The probability distribution is given by

1 (x−µ)2
f (x) = √ e− 2σ2 − ∞ < x < ∞.
σ 2π

30
The mathematical constants are: e ≈ 2.7183 and π ≈ 3.1416;
µ and σ > 0 are population parameters: mean and standard deviation.
The main facts about the normal distribution are:

R∞ 1 (x−µ)2
* √ e− 2σ2 dx = 1, i.e. the total area under the normal probability distribution is
−∞ σ 2π
R∞ −x2 √
equal to 1 (it is proved by using that e dx = π).
−∞

* the mode and the mean are = µ, and the distribution is symmetric about its mean µ. The
symmetry implies that the area to the right of µ is 0.5 and the area to the left of µ is also
0.5.

* The standard deviation is σ, and large values of σ reduce the height of the curve and
increase the spread; small values of σ increase the height of the curve and reduce the
spread.

Formally, a continuous random variable X is said to have normal distribution with mean µ and
standard deviation σ, notation X ∼ N (µ, σ 2 ), if

1 (x−µ)2
• its probability density function is f (x) = √ e− 2σ2 , and
σ 2π
Rx0 1 (x−µ)2
• the cumulative distribution function F (x0 ) for X is F (x0 ) = P (X ≤ x0 ) = √ e− 2σ2 dx
−∞ σ 2π
• Visually

f (x)

µ x0 x

1. A normal distribution curve is bell-shaped.

2. The mean, median, and mode are equal and are located at the center of the distribution.

3. A normal distribution curve is unimodal (i.e., it has only one mode).

4. The curve is symmetric about the mean, which is equivalent to saying that its shape is the
same on both sides of a vertical line passing through the center.

5. The curve is continuous; that is, there are no gaps or holes. For each value of x, there is a
corresponding value of y.

31
 6. The curve never touches the x−axis. Theoretically, no matter how far in either direction
the curve extends, it never meets the x−axis, but it gets increasingly close.
7. The total area under a normal distribution curve is equal to 1.00, or 100%.
8. The area under the part of a normal curve that lies within 1 standard deviation of the
mean is approximately 0.68, or 68%; within 2 standard deviations, about 0.95, or 95%;
and within 3 standard deviations, about 0.997, or 99.7%.
Definition 59. The standard normal distribution is a normal distribution with a mean of 0 and a
standard deviation of 1. It is denoted by Z, i.e., Z ∼ N (0, 1).

Studying the distribution X ∼ N (µ, σ 2 ) reduces to studying the distribution of Z ∼ N (0, 1)

For example, if one is to find P (X ≤ x0 ), then one computes

 
X −µ x0 − µ
P (X ≤ x0 ) = P (X − µ ≤ x0 − µ) = P ≤
σ σ
= P (Z ≤ z0 ) ,
X−µ x0 −µ
where Z = σ
and z0 = σ
.
The area under the standard normal curve to the left of z0 represents the probability that
P (Z ≤ z0 ) = P (X ≤ x0 ).

Visually, we transform

f (x)

µ x0 x

into

f (z)

0 x0 −µ z
σ

32
Example 60. Let X be a normally distributed random variable with a mean of 10 and a standard
deviation of 2. Find the probability that x lies between 11 and 13.6.

Solution: Standardizing we find

11 − 10 13.6 − 10
z11 = = .5 z13.6 = = 1.8
2 2

P (11 ≤ X ≤ 13.6) = P (.5 ≤ Z ≤ 1.8)

= P (Z ≤ 1.8) − P (Z ≤ .05) = .9641 − .6915 = .2726

Problem 61. Calculate the area under the standard normal curve between these values:

(a) z1 = −1.4 and z2 = 1.4

(b) z1 = −3 and z2 = 3

Solution: We calculate the area using MS Excel / the ALEKS calculator / the table in the
appendix

(a) The area between z1 = −1.4 and z2 = 1.4 is

P (−1.4 ≤ Z ≤ 1.4) = P (Z ≤ 1.4) − P (Z ≤ −1.4) = 0.9192 − 0.0808 = 0.8384 .

(b) The area between z1 = −3 and z2 = 3 is

P (−3 ≤ Z ≤ 3) = P (Z ≤ 3) − P (Z ≤ −3) = 0.9987 − 0.0013 = 0.9974 .

Problem 62. Find the following probabilities for the standard normal random variable Z:

(a) P (−1.43 < Z < 0.68)

(b) P (.58 < Z < 1.74)

(c) P (−1.55 < Z < −.44)

(d) P (Z > 1.34)

33
 (e) P (Z < −4.32)

Solution: We use MS Excel / the ALEKS calculator / the table in the appendix:

(a) P (−1.43 < Z < 0.68) = P (Z < 0.68) − P (Z < −1.43) = 0.7517 − 0.0764 = 0.6753;

(b) P (.58 < Z < 1.74) = P (Z < 1.74) − P (Z < 0.58) = 0.9591 − 0.7190 = 0.2401;

(c) P (−1.55 < Z < −.44) = P (Z < −0.44) − P (Z < −1.55) = 0.3300 − 0.0606 = 0.2694;

(d) P (Z > 1.34) = P (Z ≤ −1.34) = 0.0901;

(e) P (Z < −4.32) ≈ 0.

Problem 63. Find these probabilities for the standard normal random variable Z:

(a) P (Z < 2.33)

(b) P (Z < 1.645)

(c) P (Z > 1.96)

(d) P (−2.58 < Z < 2.58)

Solution: We use MS Excel / the ALEKS calculator / the table in the appendix:

(a) P (Z < 2.33) = 0.9901;

0.9495 + 0.9505
(b) P (Z < 1.645) = = 0.95;
2
(c) P (Z > 1.96) = P (Z ≤ −1.96) = 0.0250;

(d) P (−2.58Z < 2.58) = P (Z ≤ 2.58) − P (Z ≤ −2.58) = 0.9951 − 0.0049 = 0.9902.

Problem 64. (a) Find a z0 such that P (Z < z0 ) = 0.025

(b) Find a z0 such that P (Z < z0 ) = 0.9251

Solution: To find such z0 we again use MS Excel / the ALEKS calculator / the table in the
appendix

(a) For P (Z < z0 ) = 0.025 such z0 is z0 = −1.96, since P (Z < −1.96) = 0.025.

(b) For P (Z < z0 ) = 0.9251 such z0 is z0 = 1.44, since P (Z < 1.44) = 0.9251.

Problem 65. Find a z0 such that P (−z0 < Z < z0 ) = 0.8262.

34
Solution: If z0 is such that P (−z0 < Z < z0 ) = 0.8262, then we should have

0.8262 = P (−z0 < Z < z0 ) = P (Z < z0 ) − P (Z < −z0 ) = 1 − 2 × P (Z < −z0 ) ,

1 − 0.8262
hence P (Z < −z0 ) = = 0.0869, from here −z0 = −1.36, therefore
2
z0 = 1.36 .

Problem 66. (a) Find a z0 that has area .9505 to its left.

(b) Find a z0 that has area .05 to its left.

Solution:

(a) If z0 is such that the area to the left of it is 0.9505, then P (Z ≤ z0 ) = 0.9505, hence z0 = 1.65.

(b) If z0 is such that the area to the left of it is 0.9505, then P (Z ≤ z0 ) = 0.05, hence z0 =
−1.645.

Problem 67. (a) Find a z0 such that P (z0 ≤ Z ≤ z0 ) = 0.90.

(b) Find a z0 such that P (z0 ≤ Z ≤ z0 ) = 0.99.

Solution:

(a) If z0 is such that P (−z0 < Z < z0 ) = 0.90, then we should have

0.90 = P (−z0 < Z < z0 ) = P (Z < z0 ) − P (Z < −z0 ) = 1 − 2 × P (Z < −z0 ) ,

1 − 0.90
hence P (Z < −z0 ) = = 0.05, from here −z0 = −1.645, therefore
2
z0 = 1.645 .

(b) If z0 is such that P (−z0 < Z < z0 ) = 0.99, then we should have

0.99 = P (−z0 < Z < z0 ) = P (Z < z0 ) − P (Z < −z0 ) = 1 − 2 × P (Z < −z0 ) ,

1 − 0.99
hence P (Z < −z0 ) = = 0.005, from here −z0 = −2.575, therefore
2
z0 = 2.575 .

35
5.3 Applications of the normal distribution
Problem 68. A publisher has discovered that the number of words contained in new manuscripts is
normally distributed, with a mean equal to 20,000 words in excess of that specified in the author’s
contract and a standard deviation of 10,000 words. If the publisher wants to be almost certain (say, with
a probability of 0.95) that the manuscript will have less than 100,000 words, what number of words
should the publisher specify in the contract?

Solution: Let X be the number of words in excess of that specified in the author’s contract. From
the given information we have that X is a random variable with normal distribution for which
µX = 20000 and σX = 10000. The task is to find x0 such that P (X + x0 ≤ 100000) = 0.95. We
have

0.95 = P (X + x0 ≤ 100000) = P (X ≤ 100000 − x0 ) =

   
X − µX 100000 − x0 − µX 100000 − x0 − 20000
=P ≤ =P Z≤ =
σX σX 10000
 
80000 − x0
=P Z≤ ,
10000
80000 − x0
hence = 1.645, and thus x0 = 80000 − 16450 = 63550. Thus the publisher should
10000
specify 63550 words in the contract if he wants to be 95% certain that the manuscript will not
exceed 100,000 words.

Problem 69. For a car traveling 30 miles per hour (mph), the distance required to brake to a stop is
normally distributed with a mean of 50 feet and a standard deviation of 8 feet. Suppose you are traveling
30 mph in a residential area and a car moves abruptly into your path at a distance of 60 feet.

(a) If you apply your brakes, what is the probability that you will brake to a stop within 40 feet or
less? Within 50 feet or less?

(b) If the only way to avoid a collision is to brake to a stop, what is the probability that you will avoid
the collision?

Solution: Denote by X the distance traveled by a randomly selected car going with 30 miles
per hour (mph), until it comes to a stop. Given is that X is normally distributed with µX = 50
and σX = 8.
40 − 50
(a) The probability that the car stops within 40 feet or less is P (X ≤ 40) = P (Z ≤ )=
8
P (Z ≤ −1.25) = 0.1056. The probability that the car stops within 50 feet or less is P (X ≤
50 − 50
50) = P (Z ≤ ) = 0.5.
8
(b) If the only way to avoid a collision is to brake to a stop while traveling 30 mph in a resi-
dential area and a car moves abruptly into our path at a distance of 60 feet, the probability
to avoid the collision will be
60 − 50
P (X ≤ 60) = P (Z ≤ ) = P (Z ≤ 1.25) = 0.8944 .
8

36
Problem 70. Suppose that you must establish regulations concerning the maximum number of people
who can occupy an elevator. A study of elevator occupancies indicates that if eight people occupy the
elevator, the probability distribution of the total weight of the eight people has a mean equal to 1200
pounds and a standard deviation of 99 pounds. What is the probability that the total weight of eight
people exceeds 1300 pounds? 1500 pounds? (Assume that the probability distribution is approximately
normal.)

Solution: Let X be the total weight of 8 people. It is given that X has normal distribution
with µX = 1200 and σ = 99. The first question is about P (X > 1300) and the second is about
X − µX
P (X > 1500). Since Z = will be the standard normal variable, we find
σX
1300 − µX 1300 − 1200
P (X > 1300) = P (Z > ) = P (Z > ) = P (Z ≤ −1.01) = 0.1562, and
σX 99
1500 − µX 1500 − 1200
P (X > 1500) = P (Z > ) = P (Z > ) = P (Z ≤ −3.03) = .0012 .
σX 99

Problem 71. A grain loader can be set to discharge grain in amounts that are normally distributed,
with mean µ bushels and standard deviation equal to 25.7 bushels. If a company wishes to use the loader
to fill containers that hold 2000 bushels of grain and wants to overfill only one container in 100, at what
value of µ should the company set the loader?

Solution: Denote by X the amount of bushels loaded in an arbitrarily chosen container. We are
given given that X is normal with unknown µX = µ and σX = 25.7. As it is required that only
one in 100 containers, each with capacity 2000, gets overfilled, we must have 0.01 = P (X >
2000 − µX µ − 2000 µ − 2000
2000) = P (Z ≥ ) = P (Z ≤ ), hence = −2.33 and thus
σX 25.7 25.7
µ = 2000 − 2.33 × 25.7 = 1940.119 .

Problem 72. A machine operation produces bearings whose diameters are normally distributed, with
mean and standard deviation equal to 0.498 and 0.002, respectively. If specifications require that the
bearing diameter equal 0.500 inch ± 0.004 inch, what fraction of the production will be unacceptable?

Solution: Denote by X the diameter of a randomly selected bearing. Given is that X has
normal distribution with µX = 0.498 and σX = 0.002. The unacceptable bearings will be those
for which X < 0.5 − 0.004 = 0.496 or X > 0.5 + 0.004 = 0.504. Then the fraction of unacceptable
will be
0.496 − µX 0.504 − µX
P (X < 0.496) + P (X > 0.504) = P (Z ≤ ) + P (Z ≥ )=
σX σX
0.496 − 0.498 0.504 − 0.498
= P (Z ≤ ) + P (Z ≥ )=
0.002 0.002
= P (Z ≤ −1) + P (Z ≥ 3) = 0.1587 + 0.0013 = 0.16 = 16% .

37
Problem 73. Although faculty salaries at colleges and universities in the United States continue to
rise, they do not always keep pace with the cost of living nor with salaries in the private sector. In 2005,
the National Center for Educational Statistics indicated that the average salary for Assistant Professors
at public four-year colleges was $50581. Suppose that these salaries are normally distributed with a
standard deviation of $4000.

(a) What proportion of assistant professors at public 4-year colleges will have salaries less than $45000?

(b) What proportion of these professors will have salaries between $45,000 and $55,000?

Solution: Let X be be the salary of a randomly chosen Assistant Professors. It is given that X
is normally distributed with µX = 50581 and σX = 4000. Then the calculation is easy:

(a) The proportion of assistant professors with salaries less than $45000 is

45000 − µX 45000 − 50581

P (X ≤ 45000) = P (Z ≤ ) = P (Z ≤ ) = P (Z ≤ −1.39525) = 0.08155 .
σX 4000

(b) The proportion of professors with salaries between $45,000 and $55,000 is

P (45000 ≤ X ≤ 55000) = P (X ≤ 55000) − P (X ≤ 45000) =

55000 − µX 45000 − µX
= P (Z ≤ ) − P (Z ≤ )=
σX σX
55000 − 50581 45000 − 50581
= P (Z ≤ ) − P (Z ≤ )=
4000 4000
= P (Z ≤ 1.10475) − P (Z ≤ −1.39525) =
= 0.8654 − 0.08155 = 0.7839
≈ 78% .

References
[Blu2017] Alan G. Bluman: Elementary Statistics: A Step by Step Approach, 10th Ed. (McGraw-
Hill) – an electronic textbook integrated in McGraw-Hill ALEKS 360, 2017.

38