Lecture

interested in
Population : The collection of individuals we are
learning
about

Sample : The collection of individuals we

get to see as a subset of

variable observed subset

population OR a collection of ransom as a

of
population
Population Parameters and sample Statistic
Population Parameter Sample Statistic
sample mean

3
>
Population Mean-e
-

sample Standard
Population Standard-g deviation
-

S
deviation 1
Point Estimators
sample proportion p
-

P
Population Proportion
-

Sample correlation
correlation coefficient Pxy =

rxy
coefficient
Statistic -

Any function of our

sample/data connected

parameter
-
A certain characteristic of a
population
Definitions -
based on
experiments
-

Sample Space(s) : set of all

possible outcomes of an
experiment
Examples'
1) S = Eall possible outcomes 3 H-Heads
T-Tail
S =
EH T3 ,
=

Two
possible outcomes
 times List sample
2) A fair coin is flipped 3 consecutive .
space .
H H) ,

E 3
(H, H H) , ,
(T , ,

S =
CH , H T) , , (T , H ,
T) ,

(H ,
T, H) , (T , TcH) ,

(H , +, T) , (t , + ,T)

3) Record the wait time when calling customer service .

S =
Gall possible waiting times 3 =
[0 ),

or subset of
outcomes of interest
-

Event-collection of ,

(S , H) ,
&CH 3
H)
, ,
Flip two coins S =
(H , )) , (S , S)

-
Ex 1 : First coin flip is heads

A =
ECH ,
H) ,
(H , +13

-
Ex 2 : Exactly one head
A =
ECH , S) ,
(S , H)3
-

Ex3 : At least I Tails

A =
&(S , H) ,
(H , S)
,
15 , 3)3
Venu Diagrams
Sisumple Space)
y
A x B
*

O
% O
Ex: consider tossing a single (Six-sided) Gie .

what is sample space ?

S = E1 ,
2
, 3,4 , 5
, 63
Consider events :
-

three

A = [ 1 , 23 ,
B =
[1 33 ,
c =
52 ,
4
, 63
,

T
Roll 2
↑ ↑
I or
Roll Roll an even number
1 or 3

-complements ; The complement of an event A, denoted by Al , is the set

of all outcomes in S that aren't contained in A

.

Recall A : E1 , 23
-

A =
23 ,
4 ,
51 63
-

Intersection : Denoted by ARB and read "A and B" -

is the event
,

consisting of all outcomes that are in both , A and B,

Recall A =
& 1 23 ,
,
B = E1 33,

AnB =
213
-

Union : Denoted by AUB and read "A B"

-

or

-
Recall A =
31 , 23 ,
B= 21 , 33
-

AUB =
E1 , 2 33 ,

Empty Set : Denoted by 0

-

Recall A =
3 1 , 27 ,
B = [1 , 33 ,
C =
32 ,
4
,
63
-
B1 =
23 =
0
Two sets A and B are mutually exclusive or disjoint if their intersection
is in the . In other words , A B have elements in
empty set and no common ,

Recall A E1 , 25 63
=
B El 33 &L
-

=
, ,
,
c =
,
4
,

B1 = E3 =
0 Band Care disjoint
Lecture 2

probability function
P assigns probabilities to any event AL

P : Maps probabilities to events

I! Any event

Output ; Probability of that event

Long Run Relative Frequency ; proportion of times an event occurs if

experiment is
repeated an infinite amount of times

# an event occurs
Relative Frequency =

# experiment is run

Probability Axioms

1) For event A , PCA) 20 (should be nounegative)

2)p(s) =
1(p(0) = 0)

3) For disjoint infinite events Al , Az , ...

PCA ,
or A <
or ... )
= PLA
Examples ;
Color Brown Red Yellow Green Tan
Orange
Probability 0. 3 0 2
.
ar 0. 1 0 1
.
?

Probability that it is Tan is .

0

If there are N# of
equally likely
-

outcomes
,
the prob for each
one is
N

P(A)) = 1 -

P(A)

Addition Rule ; For any two events A and B,

-

For disjoint events :

PLAUB) = P(A) + P(B)

-
For non-disjoint events :

P(AUB) = PLA) + P(B) -PLANB)

Examples : Pick a card from a standard deck; A is draw a heart and

B is draw a face car

P(ANB)
52 cards , 4 suits , each suit has 12 cards

B
 -

PLAUB)

PCAUB) = PLA) +
O(B) -

PLANB)

=+ * -

= 2
52

P(B)

=
1-E =

-PLAND) =B
Addition Rule for three events :

PCAUBUC) =
P(A) + P(B) + P(c) PLANB) PLANC) PCB1C)
-

+ PLANBC)
-

ex : prob that student is

sophomore is ,
20 ,
prob that student is cheme
Major is 14 36 % of students are sophomores or chem e
majors
,
What
.

is prob that a student is a sophomore majoring in cheme ?

Y = is sophomore
C =
cheme major

P(YOCs P(Y) + P(C) PLYMC)

-

P(Y1C)
=

P(7) =, 20 ·
36 =. 28 +, 14 -

P(Y1C)
P(c) =.. 14 P(Y1C) =, 42- ·
36
36 =. 06
PlYUC) =
Binary Outcomes

W= success

W= F = Failure

W and F disjoint
-

are

IF pCul = p ,
what is PCF1 ?

P(F) = 1 -

PLA) =. 12 P(A2) =. 0 > P(As) =. 05

PCA , VAz) =. 13 PLA UAs) ,

=. 14

PlAcUAs) = I PLAINAzRAs) =. Ol

11 Whats prob that system doesn't have A ,?

PCA)) = 1 -

P(A)

=, 88

2) Whats that system has both A and Ar ?

prob ,

=, 06
P(A MA2) = 12 +.
,
07- 13 ,

3) Whats prob that system has both A, and Az but not As ?

PLAINAzRAs)
PLA , MAr) = PLAinArNAs) + P(A , nAudAs)

=>. 06 = PLANAr1As)) +, 0

PlAinArNAs') =
,
05
Lecture 3

Product Rule for Ordered outcome

-

Mu rule States that with in elements and n elements

a , an ,
...

b
, by it's possible to form run
...,
pairs

Example 1 ; Tossing pair of dice

6 . 6 = 36 outcomes

Example 2 : Roll 2 fice and flip I coir

6 002
.
= 72 outcomes

ex : iPhone allows 4-digit pins. How many can be formed ?

pins
10 numbers> 10" =
10000
pins
4
digits
How many if other number is off ?
every

5 10
5

% 10 =
2500
. ·

permutation
ex : Club wants to fill 3 positions from a pool of n = 10
applicants
An ordered arrangement of distinct objects is called a
permutation
10 :

9 . 8 => 20 outcomes

so nocu-1) ·
(n -

2) ..... (n -
r+ 1) =
c) :
 is be selected 25
.
ex : Committee of 3 to from

25
. 24- 23 =
E = 1150 permutations

ex : Task force of 3 with

people same rank from pool of n = 10

= >20
1009 8 .

permutations

An unordered arrangement r
of distinct objects is called a combination

Multiple Combination :

15 faculty cross-listed How

Suppose 6 of are .
many ways can a

committee of 3 be formed if at most 1 is cross-listed ?

If I cross-listed :

(9) (2) ·
=
so = 1026

If o cross-listed :

(i) = =
969 combinations

1026 + 969 = 1995 combinations

ex : 4 of 25 faculty from stats are cross-listed in Math.

It runtomly choose 5 people , whats prob that it contains I members

cross-listed ? Round to 2 decimal
that
places
are

Isl = N =
(4 =
53130

(2)(3) < 500 =

PLA) =
-8 0 .
15
Lectura 4 & Conditional
Independence Probability
-

Conditional Probability - Prob of event that another event has

an
given
already occured

PLAIB)
ABS
= = Prob of A give a

Ex : 200 selected
people Total
Republican Democrat Interpendent
40 15 118
College 55

40 20 38 90
No college
Total 95 60 45 20 %

P(Republican) =
1 = 0 475
.

P(College) = = 0 55
.

prepublican /College) ==
P(No College ( Independent) = =2
455

=3 Using the definition

of conditional probability

General Multiplication Rule

PLANB) =
PLAIB) P(B) -

PCAID) =

POBIA PAT
Examples : Find ARB

1) A =
Efirst roll even 3 ,
B =
35 on 2nd roll3
panB) =
7 5-
=
+
2) A=
&2 ,
3, 4 1st roll 3 ,
B = 5066 on Lnd roll3
PLANB) == =
I
3) A = E1 on 1st roll3 , B = E6 on 2nd roll3
PLANB) =
to % = 5
Draw three cards another what is that all 3
one after
prob cards

diamond ?

PLPPRP) = =E
Lecture 5 Prob trees & Total
probability
-

P(ANB) 8 = 0 72
BIA
= 0 9 .
: 0 . .

0
8 A 08
- B'IA PLAnB') = 0
.

-
.

S BIA' PLANB) = 0, 08
-
A
0 . 2
B'IA' PLA'nB1) = 0 12
.

One successful harvest" 0 . 16

least I successful , 88
harvest : 0
At

Suma outcome : 0 84
Both .

least I unsuccessful : 0 .
28
At
 of
Law Total Probability
PLA)
= PLANEil
= PLAE PlEl
Ex : 30 % iPhone 70 % Android .
80 % iPhone and 90 % Android game.
,

whats overall % of students who game on

phone ?
8 + 07 0 % 0 87
p(b) 0 =
= 0 3 .
.
.
.
% . .

50 %. Whats make
6 : 30 am 80 %. Weekends
6 : 30am prob
Weekdays make
up

up by 6 : 30 am ?

(0 8)(E) + (0 5)(E) = 0 71.

P(E)
.

= .

Bayes Theorem

Ex: Prob without disease tests negative = 0 .

9
person .

Prob disease 0 95.

person with tests positive = .

In 8% have disease.
population
0 95 = 0 168
P(T) = 0 08
. .
.

Prob that person has disease it test is

positive ?

TDPDip(D) o
=
0 452
.

02
Lecture 6 -
Intro to Random Variables

Discrete Ransom Variable -

A variable that can

assume only a finite

number of distinct values

random variable Y is
Probability distribution for a discrete a list of

each for Y with probability of observing that value in

possible value

an
experiment P(Y .
=
y) or
ply)
p() = P(y = 1)

ply) is also called the probability Mass function (pot)

ply) fully describes the prob distribution of Y

P(Y = y) is defined as the sum of the

probs of all sample points in
that are assigned to the value of y
.

# heads J SCH H CHit) , CT,H ) CT . T

eg : Suppose 2 coins are
flipped , y
=
,
= . , ,

=
p(Y 1) =
P( & (H , (3) + p(3CT , H(3)
=
p()
For any prob dist of a discrete random variable y ,
the two conditious are met:

1) 0 [ P(Y = ) =
p(y) 11 Xaxioms 1 and 2 of probability
2)
Ep(y) = 1 * axious 2 and 3 of
probability
all y

ex : Toss coin 3 times and let y be # of heads .

Write out and plot dist

for this random variable .

Y y p(y)
Hit It ? O
j
HH T
S = 2
H + H

H + T I
158-
THI
THT
T 28 = -

TTH
! 3 -
TTT
oi !
ex : Urn contains 6 balls ,
3 red and 3 green . Choose I balls randomly without

replacement .
Y is # of red balls .

13 (E) = :
= 20 combinations

2) (b) ways
to choose y red balls and Im) ways to choose 3 -

green balls

3)
(i) (s 3)
with Y=

outons as
·
outcomes

4) p(y)
=
ps
=
)
=
5)4p(y)
plastic plast -
o
:
I

2il to =

>
0 123
Lecture 7 -
Discrete Random Variables and CDF

X =
So Aorr
Random variable X is called a Bernoulli random variable

Suppose p
= 0 1 .

,
the put of the Bernoulli r X will be
1
p(o) = 1 0 = 0 4
-

.
,

0. 1
p(1) =

= P
p(1)
= 1 P
p(o)
-

Cumulative Distribution Function (CDF)

F(x) =
p(X = x) =
G p(y)
4 : 41X

Properties'
1) F(x120 X

2) F is non-decreasing ,
i . e , x , < X2 implies that F(x ) [F(x2)
,

3) F is normalized : Lim F(x) = 0 and lim F(X) =

X - X -x
-
X

4) F is right continuous

is by :
beds suppose part given
of X
ex ! Let X be # of
occupied ;

X0
123
2 0 25 0 3 0 25
p(x) 0 .
. .
.

X(0

E
0
, ·
1-
0 . 2
, 02X(/ 0 3-
75 - ·o
12xch
.

45 0
F(x)
0 .

= , ,

402-
1 oo
0
122x3 0 .
45-

. , 32x ,
0 2- .
0 .
15-0

· iii 3
 > 1346

p(y) 0 20
.
.
40 1 . 0 3
.

E
y2

F(y) = P(Y(y) = 0
12
32414
n2y16
2
. ,

,
624
the Cof find P(Y 3) :
Using
=
,

4
P(Y = 3) = F(3) -
F(2) = 0 .
6 -0 .
2 = 0 .

P(Y y) =

=
E I

Find F(4)"
2) 2
F(u) = P(Y(4)= p(Y = = 0 .

Find P(62x [10 5) .

8:
P(y = y)
E
=

0
4
, = 10
.

y
0 . 1 , y = 11

4 = .
0
3 +0
X = 10) P() + p((0) = 0 .

< 10 5)
P(X =) or = .

=
P(62x .