INTRODUCTION TO

STATISTICS I

CHAPTER 4

1
INTRODUCTION TO SET THEORY
English Logic Set Theory
p,q,r P,Q,R
And p  q P Q
Or p  q P  Q
Not ~p P or P c
Implies pq P Q

2
INTRODUCTION TO
PROBABILITY

PROBABILITY is a numerical measure

between 0 and 1 that describes the chances or
likelihood that an event will occur.
 Probabilities closer to 1 indicate that the event is highly
likely to occur.
 Probabilities greater than 0.5 indicates that the event is
more then likely to occur
 Probabilities closer to 0 indicate that the event is
unlikely to occur.

3
CERTAINTIES AND IMPROBABLE

SURE EVENT: if the probability of an

event is 1, then the event is certain to occur
IMPOSSIBLE EVENT: if the probability
of the event is 0, then the event is
improbable; that is, the event is certain not
to occur.

4
NOTATION
P for " Probabilit y"
( for " of"
E for the " event"; also used A, B, C, ...
) denotes the end of the event
 " is"
n for the " number" of outcomes decribed by the event
S for the " sample space" , the entire set of possible outcomes

5
FORMULATION OF PROBABILITIES
IN A DISCRETE MEASURE

“the probability of an event is the number of

P ( E)  n (
ways an event happens compared to the total number

E)  n
of possible outcomes in the sample space”.

( S)
6
FORMULATION OR
TRANSLATION

n( E )
P( E ) 
n( S )

7
 ALTERNATIVE
FORMULATION
This measure of part to how in a sample is called
the relative frequency

f
P( E )  ;
n
f  n( E ) and n  n( S )
8
LAW OF LARGE NUMBERS
(LLN)

In general: As the sample size increase, the statistic

approaches the parameter
In terms of Means: As the sample size increase, the
sample mean approaches the population mean
In terms of Proportions: As the sample size increase, the
sample proportion (relative frequency) approaches the
population proportion (true probability)
In terms of Variance: As the sample size increase, the
sample variance approaches the population variance
ETC.

9
LLN IN SHORTHAND
Let  represent the parameter and
ˆ represent the statistic; then...
In general, as n  , ˆ  
For Means, as n  , x  
f
For Proportions, as n  , pˆ  p
n
For Variance, as n  , s 2   2

10
TERMINOLOGY
A STATISTICAL EXPERIMENT or
STATISTICAL OBSERVATION as any random activity that
results in a definite outcome
An EVENT is a collection of one or more outcomes of a
statistical experiment
A SIMPLE EVENT is an outcome of a statistical experiment
that consists of one and only one of the outcomes
The set of all simple events constitutes THE SAMPLE
SPACE of an experiment
Two events are MUTUALLY EXCLUSIVE if there
intersection is empty.

11
PRINCIPLES OF
PROBABILITY
The probability of the sample space is one.
 Mathematically: P(S )  1
The probability of an event is at least zero and at most one.
 Mathematically: 0  P( E )  1
Given SIMPLE EVENTS, that is, if events are MUTUALLY
EXCLUSIVE and in UNION FORM THE SAMPLE SPACE, then
the sum of the probability is one.
 Mathematically:

if E i  E j  ; i  j
and E1  E2    En  S
then P( E1 )  P( E2 )    P( En )  1 12
MUTUALLY EXCLUSIVE EVENTS

13
SET THEORY: VENN DIAGRAMS

14
COMPLEMENT

The COMPLEMENT of an event, E, is

the set of outcomes NOT in E denoted

E or E 
c

15
COMPLEMENTARY EVENTS

E  E  S
E  E   ,
where     , the empty set
and
P(E)  P(E )  1
16
EXAMPLES: PROBABILITIES
Given the three even are mutually exclusive and are the
simple events which form the sample space, determine
which of the follow probability distributions are plausible;
if the probability distribution is implausible, state why?

P( E1 )  12 , P( E 2 )  13 , P( E3 )  1
6

P( E1 )  0.145, P( E 2 )  0.521, P( E3 )  0.333

P( E1 )  12 , P( E 2 )  23 , P( E3 )   16
17
 ANSWERS: PROBABILITIES
Given the three even are MUTUALLY EXCLUSIVE and
are the SIMPLE EVENTS which FORM THE
SAMPLE space, determine which of the follow
probability distributions are plausible; if the probability
distribution is implausible, state why?
P( E1 )  12 , P( E 2 )  13 , P( E3 )  1
6
plausible

P( E1 )  0.145, P( E 2 )  0.521, P( E3 )  0.333   0.999  1, implausible

P( E1 )  12 , P( E 2 )  23 , P( E3 )   16 P( E3 )  0, implausible

18
RECAP
PROBABILITY is a measure between 0 and 1,
inclusive, which describe the chance of an event
occurring
The probability of the sample space is 1.
The probability of any event is at least 0 and at most 1
The sum of the probabilities of all simple events is 1
Mutually exclusive events are such that their intersection
is the empty set
The sum of the probability of an event and its
complement is one

19
SOME PROBABILITY RULES
FOR COMPOUND EVENTS
Two events are INDEPENDENT if the
occurrence or nonoccurrence of one event
does not change the probability that the
other event will occur
Otherwise, if there is a implied
CONDITION between the occurrence of
the two events, then the events are
dependent.
20
INDEPENDENT EVENTS
Two events are INDEPENDENT if the probability of
both A and B happening is simply the probability of A
times the probability of B. NOTE: In general, when
there are two events, AND means MULTIPLY

P( A and B)  P( A)  P( B)

21
DEPENDENT EVENTS
Two events are dependent if the probability of both A
and B happening is the probability of A times the
probability of B GIVEN A has occurred; AND still
means MULTIPLY, but a condition has been set forth

P( A and B)  P( A)  P( B given A)
OR P( A)  P( B | A)
OR P( B )  P(A |B)

22
EXAMPLE:

INDEPENDENT/DEPENDENT
Two cards are selected from a standard deck of 52
cards WITH REPLACMENT, what is the
probability that the two cards selected are Hearts.
INDEPENDENT:
P( H and H )  13
52  13
52  1
16

Two cards are selected from a standard deck of 52

cards WITHOUT REPLACMENT, what is the
probability that the two cards selected are Hearts.
DEPENDENT:
P( H and H )  13
52  12
51  1
17

23
 PROPERTIES:
AND MEANS MULTIPLY
Given two events, A and B, are INDEPENDENT
then
P( A and B)  P( A)  P( B)

Given two events, A and B, are DEPENDENT then

P( A and B)  P( A)  P( B | A)

24
CONDITIONAL PROBABILITY

Two events are dependent if the probability of both A

and B happening is the probability of A times the
probability of B GIVEN A has occurred; hence,
solving for the conditional statement we have,

P( A  B)
P ( A | B) 
P( B)
P( A  B)
P ( B | A) 
P ( A)
25
OR OUT OF THE “NEW UNIVERSE”

n ( A B )
P( A  B) n( S )
P ( A | B)   n( B)
P( B) n( S )
n ( A B ) n ( A B )
 n( S )  n( S )
n( B)  n( B )  PART
WHOLE

26
 CONDITIONAL
EXAMPLES:

PROBABILITIES
FEMALE MALE
EYE COLOR TOTAL
BROWN 20 5 25
BLUE 15 30 45
GREEN 30 50 80
TOTAL 65 85 150

27
EXAMPLES : GIVEN FEMALE
FEMALE MALE
EYE COLOR TOTAL
BROWN 20 5 25
BLUE 15 30 45
GREEN 30 50 80
TOTAL 65 85 150

P( Brown | Female)  20
65  134

28
EXAMPLES: GIVEN BROWN
FEMALE MALE
EYE COLOR TOTAL
BROWN 20 5 25
BLUE 15 30 45
GREEN 30 50 80
TOTAL 65 85 150

P( Female | Brown )  20
25  54  0.8

29
PROPERTIES: OR MEANS ADD
(MINUS OVERLAP)

Given two events, A and B, are MUTUALLY EXCUSIVE

P ( A  B )  P ( A)  P ( B )
Given two events, A and B, are NOT MUTUALLY
EXCUSIVE

P ( A  B )  P ( A)  P ( B )  P ( A  B )

30
 A

OVERLAP ILLUSTRATED

P ( A  B )  P ( A)  P ( B )  P ( A  B )
31
EXAMPLE: MUTUALLY EXCLUSIVE

Given two events, A and B, are MUTUALLY EXCUSIVE

32
EXAMPLE: NOT EXCLUSIVE
Given two events, A and B, are NOT MUTUALLY
EXCUSIVE

33
EXAMPLE:
a) Relative frequency;
19/317 = 0.0599 or 5.99%

b) 1 − 0.0599 = 0.9401
or approximately 94%

c) Defective, not defective;

c) D or D the probabilities add up to
one.
P( D)  P( D)  1
0.0599  0.9401  1
34
EXAMPLE:

a) With replacement, P(Red first and Green second) =(3/12)(7/12) =

7/48 or 0.146
INDEPENDENT

b) Without replacement, P(Red first and Green second) = (3/12)

(7/11) = 21/132 or 0.159
DEPENDENT
35
EXAMPLE: (a) Not satisfied
P(NS) = 35/360
(b) Not satisfied and walk-in
P(NS and WI) = 21/360
(c) Not satisfied, given referred
P(NS|R) = 5/149
(d) Very satisfied
P(VS) = 107/360
(e) Very satisfied, given referred
P(VS|R) = 48/149
(f) Very satisfied and TV ad
P(VS|TV) = 31/360
(g) Are the events satisfied and
referred independent or not?
Explain your answer.
No; P(referred) = 149/360 is
not equal to P(referred, given
satisfied) = 59/138.

36
TREES AND
COUNTING TECHNIQUES
MULTIPLICATION PRINCIPLE:
 AND means MULTIPLY

SUMMATION PRINCIPLE;
 OR means ADD

37
EXAMPLE: MULTIPLICATION PRINCIPLE

QUESTION: Given there are three choices for

salad, five choices for the entrée and two
choices for dessert, how many dinning
selections are there for this three course
meal?
ANSWER: 3x5x2=30

38
TREE DIAGRAM: TWO FAIR COINS
ARE TOSSED (A NICKEL AND A DIME)
2  2  4 different outcomes

Nickel Dime
H
H
T
H
T
T 39
SAMPLE SPACE: TWO COINS

S   HH , HT , TH , TT 
Since these were fair coins, they are equally likely
to occur and the probabilities can be taken from the
sample space above; that is,
P( HH )  P( HT )  P(TH )  P(TT )  1
4
In general, however this is not true.

40
TREE DIAGRAM: TWO UNFAIR COINS
ARE TOSSED (A NICKEL AND A DIME)

Nickel Dime
0.2 H
0.4
H
0.8 T
0.2 H
0.6
T
0.8 T 41
SAMPLE SPACE: TWO COINS

S   HH , HT , TH , TT 
Since these were UNFAIR coins, they are not
equally likely to occur; that is,
P( HH )  0.4  0.2  0.08
P( HT )  0.4  0.8  0.32
P(TH )  0.6  0.2  0.12
P(TT )  0.6  0.8  0.48 42
FACTORIALS
Factorials count the number of ways to order n
objects WITHOUT repetition, and is denoted

n! n  (n  1)  (n  2)    (2)  (1)

1! 1
0! 1
There is always one way to do nothing – don’t do
anything. It is called the EMPTY SET
  
43
EXAMPLE: FACTORIALS
QUESTION:How many ways can someone
with OCD rearrange 5 books on a shelf?

ANSWER: 5!=5x4x3x2x1=120

44
PERMUTATIONS
This is the number of ways to arrange in
order n distinct objects taken r at a time
without repetition

45
PERMUTATIONS
What if we don’t what to order all, only
some; say r<n, then when should we stop?

P(n, r ) n Pr  n  (n  1)   (?)

46
PERMUTATIONS: EXTENSION
AND MEANS MULTIPLY

P(n, r ) n Pr  n (n  1)  (?)

  
1st 2 nd r th

P(n, r ) n Pr   n  1  1 (n  2  1)  (n  r  1)

  
1st 2 nd r th

47
EXAMPLE: PERMUTATIONS
QUESTIONS: How many ways can the first
three horses out of five cross the finish
line?

ANSWER: 5P3=5x4x3=60

48
PERMUTATIONS
FORMULATED

What if we ordered all n objects and then

removed the (n-r) objects not needed. That
is,
P(n, r ) n Pr  n  (n  1)   (n  r  1)  (n  r )   (1)
( n  r )(1)

n!
P(n, r ) 
(n  r )!

49
PERMUTATION FORMULA

n! Order all n objects

P(n, r )  Remove
(n  r )! (n-r) objects not needed

50
COMBINATIONS
This is the number of ways to combine
without order from n distinct objects
taken r at a time without repetition

51
COMBINATIONS
What if we don’t what order in addition to
only wanting say r<n objects; this is

n
C (n, r ) n Cr   
r

52
COMBINATIONS FORMULATED

Since there are r! ways to order the r

objects selected using permutations n
objects permute r objects. That is,

P (n, r )
C ( n, r ) 
r!

53
COMBINATION FORMULA

n! Order all n objects

C ( n, r )  Remove both
(n  r )!r! (n-r) objects not needed
And the order on the
remaining r objects

54
FOUR COUNTING PRINCIPLES
  REPETITION ALLOWED REPETITION NOT
ALLOWED

ORDER Multiplication Principle FACTORIALS count the

AND means MULTIPLY number of ways to Permute
  all n objects
If there are n ways to do one event  n!=n(n-1)…(2)(1)
and m ways to do a second event;  
then there are nm ways to do both events PERMUTATIONS
To count the number of ways
to permute r out of n objects
P(n,r)= n! .
(n-r)!

NO OR NOT ON EXAM COMBINATIONS

PARTIAL For example: how many distinct four-letter To count the number of ways
ORDER “words” can be made from the word “noon”; to combine r out of n objects
here repetition is allowed since there are C(n,r)= n! .
duplicate letters; however if these letters are r! (n-r)!
reversed, this is not a new distinct “word.”
 
Hence, there are 4! Ways to order the four
letters, but we must remove the order on the
n’s and the o’s; that is, divide by 2! And 2!
 4! = 6.
2! 2!
{noon, nnoo, nono, oonn, onon, onno}
55
EXAMPLE
QUESTIONS: Howmany way can you
PERMUTE 3 out of 4 letters?
Letters: L={A,B,C,D}
ANSWER:
4x3x2 = 24
ABC, ACB, BAC, BCA, CAB, CBA
ABD, ADB, BAD, BDA, DAB, DBA
ACD, ADC, CAD, CDA, DAC, DCA
BCD, BDC, CBD, CDB, DBC, DCB

56
EXAMPLE
QUESTIONS: How
many way can you
COMBINE 3 out of 4 letters? Letters:
L={A,B,C,D}
4x3x2 = 4
3x2x1
PERMUTATIONS  COMBINATIONS
ABC, ACB, BAC, BCA, CAB, CBA  ABC
ABD, ADB, BAD, BDA, DAB, DBA ABD
ACD, ADC, CAD, CDA, DAC, DCA ACD
BCD, BDC, CBD, CDB, DBC, DCB  BCD
57
EXAMPLE*
QUESTIONS: How many way can you make
DISTINCT WORDS can you create from
the word “miss”

4! = 12
2!
miss msis mssi imss isms issm
smis smsi sims sism ssim ssmi

*This counting technique will not be on the exam

58
EXAMPLE: WORKBOOK PART 4 #7
x
n( E )  x and P ( E ) 
y
n( E )
 n( S )  y since by definition P( E ) 
n( S )
 n( E )  y  x since n( E )  n( E )  n( S )
yx n( E )
 P ( E )  
since P ( E )  1  P ( E ) 
y n( S )
and
O( E )  x : ( y  x) or x
yx

O( E )  ( y  x) : x or yx
x

NOTE : Odds can not be written as a mixed fraction

or decimal; that is, Odds must be a ratio

59
ODDS VERSUS
PROBABILITIES

60
EXAMPLE: WORKBOOK PART 4 #8
a
n( E )  a and P( E ) 
ab
n( E )
 n( S )  a  b since by definition P ( E ) 
n( S )
 n( E )  b since n( E )  n( E )  n( S )
b n( E )

 P( E )  
since P ( E )  1  P ( E ) 
ab n( S )
and
O( E )  a : b or ba
O( E )  b : a or ba
NOTE : Odds can not be written as a mixed fraction
or decimal; that is, Odds must be a ratio

61
EXAMPLE:
QUESTION: There are eight words to be matched to
their respective definitions. How many different
ways can you match the words to a definition (not
necessarily correctly so)?
ANSWER: Question 1 can be one of eight, Question 2
therefore would only be one of seven, etc. Hence the
various number of distinct orderings is
8x7x6x5x4x3x2x1=40320

62
EXAMPLE:
QUESTION: There are eight words to be matched
to five respective definitions. How many different
ways can you match the words to a definition (not
necessarily correctly so)
ANSWER: Question 1 can be one of eight, Question
2 therefore would only be one of seven, etc. but only
up to five questions. Hence the various number of
distinct orderings is 8x7x6x5x4=6720

63
EXAMPLE:
P(Profit in the First Year )  P( A)  0.65
P(Profit in the Second Year )  P( B)  0.71
P( B / A)  0.87
a ) P ( A)  0.65
b) P ( B)  0.71
c ) P( B / A)  0.87
d ) P ( A  B )  P( A)  P( B | A)
 0.65  0.87  0.57
e) P ( A  B)  P ( A)  P ( B)  P ( A  B)
 0.65  0.71  0.57  0.79

f ) P  A  B    1  P( A  B )  1  0.79  0.21
 
64
EXAMPLE:
Given seven women and five men,
a) how many six member committees can be formed?
12 11 10  9  8  7
5  7  1212 C6   924  n( S )
6  5  4  3  2 1
b) how many six member committees consisting entirely of women?
7  6  5 4  3 2
7 C6   7  n( E )
6  5  4  3  2 1
c) what is the probability of a committee consisting entirely of women?
7
P( E ) 
924

65
EXAMPLE:

ADDITIONAL QUESTION
Given seven women and five men,
a) how many six member committees can be formed?
12 11  10  9  8  7
5  7  1212 C6   924  n( S )
6  5  4  3  2 1
b) how many six member committees consisting of
two women and four men?
7  6 5 4  3 2
7 C 2 5 C 4    105  n( E )
2 1 4  3  2 1
c) what is the probability of a committee consisting of
two women and four men?
105 5
P( E )  
924 44

66
EXAMPLE:
Given L is that the person is lying and
D is that the lie detector detected a lie.
P(D | L)  0.72
P(D | L)  0.07
a) P(L)  0.90 b) P(L)  0.10
 P(D  L)  P(L)  P(D | L)  P(D  L)  P(L)  P(D | L)
 0.90  0.07  0.063  0.10  0.72  0.072
c) P(L)  0.50 c) P(L)  0.50
 P(D  L)  P(L)  P(D | L)  P(D  L)  P(L)  P(D | L)
 0.50  0.07  0.035  0.50  0.72  0.36
d) P(L)  0.15 d) P(L)  0.85
 P(D  L)  P(L)  P(D | L)  P(D  L)  P(L)  P(D | L)
 0.15  0.07  0.0105  0.85  0.72  0.612
67
EXAMPLE:

68
EXAMPLE: (CONT)

69
EXAMPLE: (CONT)

70
EXAMPLE: (CONT)

71
EXAMPLE:
P(RR)=1/15
P(RB)=1/5
P(RY)=1/15
P(BR)=1/5
P(BB)=1/5
P(BY)=1/10
P(YR)=1/15
P(YB)=1/10
P(YY)=0
72
EXAMPLE:
Since each question is
equally likely, there is
only one correct sequence
out of a total of eight.

P(Correct Answers)=1/8

73
EXAMPLE:
With TWO decks of cards and drawing ONE FROM EACH
 INDEPENDENT
52  52  2704
Hence having an ACE first and a KING second is
4  4  16
16 1
P ( AK )  
2704 169
1
P ( KA) 
169
2
P ( AK  KA) 
169
74
EXAMPLE:

n(two dice) =6x6=36

n(both even) =3x3=9
P(both even) =9/36
=1/4=0.25 or 25%

75
EXAMPLE:
Let F1 be passes French 101 and F2 be passes French 102
P( F1 )  0.77
P( F2 | F1 )  0.90
Hence,
P( F1  F2 )  P( F1 )  P( F2 | F1 )
 0.77  0.90  0.693

76