PSTAT

5A

Introduction to Statistics
Instructor: Theofanis Sapatinas
MWF 10:00-10:50 (MUSICLLCH)

04/01/2013

(Lecture 1)

04/01/2013

1 / 31

General Info

Instructor

Theofanis Sapatinas
sapatinas@pstat.ucsb.edu
South Hall 5518

Office Hours

MW 11:30-13:00

Course Website

https://gauchospace.ucsb.edu/

TA

Go to your section to meet your TA

(Lecture 1)

04/01/2013

2 / 31

Lectures

A lectures plan with the corresponding sections from the book is posted
on the course website.
After each lecture, the slides will also be posted online.

(Lecture 1)

04/01/2013

3 / 31

Discussion Sessions
Classroom Session

Lab Session

Written assignments

Online applets

5% of final grade

5% of final grade

General Comments
1

Finish the assignment during the session and return it to your TA.

Grades will be posted on Gauchospace. You have one week to make

sure that grades have been posted correctly. If you want to challenge a
grade, talk to your TA.

There are no make-up discussion sessions.

(Lecture 1)

04/01/2013

4 / 31

Quizzes & Exams

- Quizzes
1
2
3
4

Friday, April 12th (10:00-10:50)

Friday, April 26th (10:00-10:50)
Friday, May 10th (10:00-10:50)
Friday, May 24th (10:00-10:50)

- Final
Monday, June 10th (08:00-11:00)

General comments
Closed books/notes. Only non-graphing calculators are allowed.
It is your responsibility to ensure you are available at the designated
times.
There are no make-up quizzes, unless you are on a sports team.

(Lecture 1)

04/01/2013

5 / 31

Grading Scheme

Classroom & Lab Assignments

Grade
10%

Quizzes

60% (i.e., 15% each)

Final Exam

30%

TOTAL

100%

(Lecture 1)

04/01/2013

6 / 31

Where do I get help?

Ask me any questions during office hours.

Ask your TAs during the weekly sessions.

Work in groups!

CLAS http://www.clas.ucsb.edu/ is great!

Approved tutors
http://www.pstat.ucsb.edu/projects/leadtatraining/tutoring.htm

(Lecture 1)

04/01/2013

7 / 31

Academic Integrity

You should not go to another classroom or lab section.

All assignments and exams should be your own. Bring your student ID at
every exam.

Cheating is a very serious offense and at a minimum will result

in your receiving an F in this class.

(Lecture 1)

04/01/2013

8 / 31

Any questions?

(Lecture 1)

04/01/2013

9 / 31

Statistics

The material of this course consists of three parts:

1

Probability
- We use probability in order to build models for the data.

Descriptive Statistics
- We summarize and visualize a dataset.

Statistical inference
- We use probability models in order to draw conclusions from datasets.

(Lecture 1)

04/01/2013

10 / 31

Statistics

(Lecture 1)

04/01/2013

11 / 31

The Literary Digest Poll

We are in 1936...
- Franklin Roosevelt is the President of the U.S. and we are preparing to
have elections.
- The country has just recovered from the Great Depression.
- The real income just started turning upwards and there are still 9 million
unemployed.
- The Republican candidate is Alfred Landon, the governor of Kansas.
- Landon is campaigning on a program of economy in government, while
Roosevelt is defensive about his deficit financing.

(Lecture 1)

04/01/2013

12 / 31

The Literary Digest Poll

The Poll
- The Literary Digest was a magazine that had an enormous prestige at
the time.
- They had called the winner in every presidential election since 1916.
- They conducted a poll in 2.4 million individuals and predicted the victory
of Landon:
Landon Roosevelt
57%
43%

(Lecture 1)

04/01/2013

13 / 31

The Literary Digest Poll

What happened?
Roosevelt won the 1936 elections by a landslide:
Landon
38%

Roosevelt
62%

The Literary Digest went bankrupt soon after.

What did they do wrong?

(Lecture 1)

04/01/2013

14 / 31

The Statisticians Objectives

Ask the right questions

Collect useful data

Summarize the data

Make decisions and generalizations based on the data

Turn the data and decisions into new knowledge

(Lecture 1)

04/01/2013

15 / 31

Introduction to Probability

(Lecture 1)

04/01/2013

16 / 31

Modeling Random Phenomena

Probability Models
Outcomes: possible results of an experiment
Sample space/ Universe/ Population: set of all possible outcomes
(denoted by S or ).
Event: set of outcomes (e.g., A in the picture)

(Lecture 1)

04/01/2013

17 / 31

Example: Fair Die Rolling

- Sample space:
= {1, 2, 3, 4, 5, 6}
- Examples of events:

(Lecture 1)

= rolling an odd number = {1, 3, 5}

= rolling an even number = {2, 4, 6}

= rolling a 5 and a 6 = {5, 6}

= rolling a 5 = {5}.

04/01/2013

18 / 31

Example: Fair Die Rolling

- If the die is fair , then the chance of getting each side is the same, i.e.
Probability (one) = Probability (two) = Probability (three)=

Probability (four) = Probability (five) = Probability (six)

or, in other words,
P({1}) = P({2}) = P({3}) = P({4}) = P({5}) = P({6}).

(Lecture 1)

04/01/2013

19 / 31

Example: Fair Coin Tossing

- Sample space:
= {H, T }

- If the coin is fair , then the chance of it landing in Tail or Head is the
same, i.e.
Probability (head) = Probability (tail),
or, in other words,
P({H}) = P({T }).
(Lecture 1)

04/01/2013

20 / 31

Probability

Assigning probabilities
If all all outcomes are equally-likely , i.e. have the same probability to
occur
P(Event) =

(Lecture 1)

# outcomes in the event

.
Total # of outcomes

04/01/2013

21 / 31

Example: Rolling a die

What is the probability that we get a 5?

What is the probability that the die lands on an odd number?

We are able to use the classical definition of probability, because all

numbers of a die are equally likely to occur, since this is a fair die.

(Lecture 1)

04/01/2013

22 / 31

Example: Rolling a die

(i) Determine the sample space, which are the possible outcomes of a die:
= {1, 2, 3, 4, 5, 6}.
(ii) Determine the events that we are interested in:
F = {get a 5} = {5}.
O = {get an odd number} = {1, 3, 5}.
(iii) Compute the probability using the definition:
P(F ) =

P(O) =

(Lecture 1)

# of outcomes in F
1
= .
Total # of possible outcomes
6

# of outcomes in O
3
1
= = .
Total # of possible outcomes
6
2

04/01/2013

23 / 31

Example: Flip a Coin Three Times

Outcomes
HHH
THH
1

HHT
THT

HTH
TTH

HTT
TTT

Probability to get 3 heads:

P(HHH) =

(Lecture 1)

1
= 0.125
8

04/01/2013

24 / 31

Example: Flip a Coin Three Times

Outcomes
HHH
THH
1

HHT
THT

HTH
TTH

Probability to get 3 heads:

P(HHH) =

HTT
TTT

1
= 0.125
8

Probability to get 2 heads:

P(Two Heads) =

(Lecture 1)

3
= 0.375
8

04/01/2013

25 / 31

Example: Flip a Coin Three Times

Outcomes
HHH
THH
1

HHT
THT

HTH
TTH

Probability to get 3 heads:

P(HHH) =

1
= 0.125
8

Probability to get 2 heads:

P(two heads) =

HTT
TTT

3
= 0.375
8

Probability to get at least 2 heads:

P(at least two heads) =

(Lecture 1)

4
= 0.5
8

04/01/2013

26 / 31

Properties of probability

1. 0 P(A) 1, for any event A.

2. P() = 1.
An event that has probability 1 is called certain.
3. P(A) = 0.
An event that has probability 0 is called impossible.

(Lecture 1)

04/01/2013

27 / 31

Complement/Opposite Events
P(A) = 1 P(opposite of A) = 1 P(Ac )

Example: Rolling a Die

A = {roll an odd number} = {1, 3, 5}
Ac = {NOT roll an odd number} = {roll an even number} = {2, 4, 6}.
P(A) = 1 P(Ac )
P(odd number) = 1 P(even number).
(Lecture 1)

04/01/2013

28 / 31

OR Rule
If A and B cannot both happen (*)
P(A OR B) = P(A B) = P(A) + P(B)

(*) A and B have no elements in common.

(Lecture 1)

04/01/2013

29 / 31

Mutually Exclusive Events

When events A and B (denoted by A B) cannot both happen (denoted
by A B = ), then they are called Mutually Exclusive Events.

Examples
1

Elections:
A = Obama elected president
B = Romney elected president

Toss a Coin
A = {get head} = {H}.
B = {get tail} = {T }.

Roll a Die
A = {get odd number} = {1, 3, 5}.
B = {get even number} = {2, 4, 6}.
(Lecture 1)

04/01/2013

30 / 31

See you on Wednesday!

(Lecture 1)

04/01/2013

31 / 31