Chapter 0: An Overview
ST2334 Probability and Statistics1
(Academic Year 2014/15, Semester 1)
Department of Statistics & Applied Probability (DSAP)
National University of Singapore (NUS)1
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 1 / 23
Outline
1 Module Information
2 A Brief Introduction To Probability
3 A Brief Introduction To Statistics
4 Set–Up Of Statistical Problems
5 Relationship Between Probability And Statistics
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 2 / 23
Module Information I
Instructors
Dr. Ho Man-Wai
Contact
(65) 6516 3953 (Office); stahmw@[Link] (email)
Office
Block S16, Level 6, Room 101
Consultation Hours
by appointment through email
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 3 / 23
Module Information II
Aim and Objective
1 Equip students with some fundamental concepts of probability & a
solid foundation in statistical theory, &
2 Provide an indication of relevance & importance of statistical theory in
solving practical problems in the real world
Syllabus
Basic concepts of probability, conditional probability, independence,
random variables, joint & marginal distributions, mean & variance, some
common probability distributions, sampling distributions, estimation &
hypothesis testing based on a normal population
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 4 / 23
Module Information III
Lecture Time & Venue
Every Monday & Thursday (from 11 Aug 2014 to 13 Nov 2014 except 22 &
25 Sep, 6 Oct 2014), 12:00nn – 2:00pm, at UTown Auditorium 3 [Room
B1-01 between Cinnamon College (USP) and College of Alice & Peter Tan]
Tutorial Time & Venue
1 Monday, 3:00pm – 3:45pm
2 Monday, 4:00pm – 4:45pm
3 Monday, 5:00pm – 5:45pm
4 Wednesday, 1:00pm – 1:45pm
5 Wednesday, 5:00pm – 5:45pm
6 Wednesday, 6:00pm – 6:45pm
Conducted at S16-06-118 (i.e., Building S16, Level 6, Room 118) starting
from 25 or 27 Sep to 10 or 12 Nov except 22, 24 Sep & 6, 22 Oct
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 5 / 23
Module Information IV
Compulsory Textbook
Introduction to Probability & Statistics for Engineers
and Scientists (4th ed.)
written by Sheldon Ross
published by Academic Press
available for download here via NUS library
Module webpage
IVLE ([Link] for lecture notes/slides, tutorial problem
sheets, suggested solutions, test scores, and so on
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 6 / 23
Module Information V
Study advice
Lecture slides/notes are to supplement but NOT to replace the
textbook
Students are urged to
attend most, if not all, lectures & tutorials
seriously read through & understand the corresponding
materials (especially the examples) in the compulsory textbook after
lectures
attempt weekly homework problems before attending tutorials, &
attempt more problems from the textbook
Assessment
Mid–semester test (on 2 Oct 2014, Thursday during lecture): 25%
End–of–semester examination (on 3 Dec 2014, Wednesday
afternoon): 75%
Both test & examination are OPEN BOOK
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 7 / 23
Introduction
Learning Outcomes
Questions to Address: © What probability is © What statistics is
about © Why we study statistics © Set–up of a statistical problem
© Relationship between probability and statistics
Concept & Terminology: © probability/chance/randomness
© uncertainty © chance/random phenomenon © probability problem © data
© dataset © variable © measurement © observation © statistical problem
© experimental unit © population © parameter © sample © statistic
Statistical concepts & methods are not only useful but indeed often
indispensable in understanding the world around us because
Life is full of uncertainty!
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 8 / 23
What is Probability? I
The idea of probability (prob), chance or randomness goes back over
300 years in the study of games of chance
The language of probability is constantly used in an informal manner
in both written & spoken contexts:
It is likely that the Dow–Jones average will increase by the end of the
year
There is a 50–50 chance that the incumbent will seek reelection
The number (#) of students who show up today at UT-AUD 3 is
probably more than 150
The odds favor a quick settlement of the strike
It is expected that at least 20,000 concert tickets will be sold
···
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 9 / 23
What is Probability? II
Under uncertainty (i.e., in any situation in which one of a number of
possible outcomes may occur), the discipline of probability provides
methods for quantifying the chances or likelihoods, associated with
the various outcomes
The mathematical theory of probability has been applied to a wide
variety of chance/random phenomena in the fields of
genetics & bioinformatics (mutations)
physics (kinetic theory of gases)
operations research (demands on the inventories of goods)
finance (volatility of a stock)
transportation (utilization rate of the road)
queuing theory (arrival rate of customers in a counter)
···
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 10 / 23
Example 1: A Probability Problem
A claim: There are > 120 students present in today’s lecture
Of course, before noon today, no one can be 100% certain whether or
not the claim is true – number (#) of students who show up is a
chance phenomenon
It depends on
how many students who enrol in the module in this semester
whether an individual student shows up for all the students who enrol in
the module (all of them are chance phenomena)
Provided that we know the chances of presence of each individual
student, one can compute the probability (prob) for the “outcome” that
there are > 120 students in theory (though this process is demanding
& costly in time)
This is called a probability problem – all the characteristics of the
random phenomenon are known & then probabilities or likelihoods of
some possible outcomes of interest are computed
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 11 / 23
Example 2: Real Life Examples of Statistical Nature
Women make 75 cents to every dollar a man makes when they work
the same job. Yes or no?
How many lung cancer cases are tobacco–related?
People predict that it is very unlikely there will ever be another
baseball player with a batting average over 400. Right?
What is the effectiveness of condom? 94%?
Do people tend to be more persuasive when they look others directly
in the eye & speak loudly & quickly?
A surprising new study shows that eating egg whites can increase
one’s lifespan. Is it true? How many more years?
There is an 95% chance that in a room with 100 people that at least
people will share the same birthday. How about here?
···
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 12 / 23
Why Study Statistics? I
1 To know how to evaluate published numerical
facts (some of them are valid; others are invalid)
that bombard you every day
manufacturers’ claims for products
results of sociological, consumer, political polls
published results of scientific research
···
2 Your profession or employment may require you to interpret the
results of sampling (surveys or experimentation) or to employ
statistical methods of analysis to make inferences in your work
to forecast sales & profit in business
to monitor product quality in engineering & manufacturing
to assist accountants in conducting audits through sampling of
accounts
···
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 13 / 23
Why Study Statistics? II
Statistics plays an important role in almost all areas of science,
business, & industry
Statistics lies at the heart of the type of quantitative reasoning
necessary for making important advances in the sciences, such as
medicine & genetics, & for making important decisions in business &
public policy
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 14 / 23
What is Statistics About?
The discipline of statistics is the science (&, arguably, also the art!) of
learning from data to help us to arrive at intelligent judgements &
informed decisions in the presence of uncertainty & variation
Data is everywhere! Data appear in everyday activities, also in a
huge volume:
quarterly sales figures
traffic volumes
percent increase in juvenile crime
survival rates for patients undergoing
medical therapy
census figures
internet usage
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 15 / 23
Example 3: U.S. Census Form
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 16 / 23
Data/A Dataset (about Internet Usage)
Age Sex Facebook a/c Weibo a/c “Connected” # of Hours/day
22 M Y Y 3.5
24 F Y N 2
18 M N N 0.5
35 M Y Y 4
.. .. .. .. ..
. . . . .
Each row corresponds to information of the same entity (in this case:
a person) under consideration
Each column contains information about a variable
Each cell contains a measurement/value about a variable associated
with an entity
Measurements in each row constitute an observation from an entity
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 17 / 23
The Discipline of Statistics is Concerned With
Collection of data
Observational studies versus Experimental designs
Analysis of data
Selection of appropriate & well-defined statistical methodologies
Interpretation of data
Retrieval of relevant information to address questions in mind
Effective communication & presentation of results relying on data
Drawing meaningful conclusions in layman’s language
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 18 / 23
Set–up of Statistical Problems I
The problem : Of interest is certain characteristic of some entities
called a parameter (e.g., average weight of students in NUS, &
proportion of students in NUS who have a a/c)
A well–defined collection of entities (called experimental units)
constituting a population of interest, which is the source of relevant
information/all measurements to address the problem
When desired information is available for the whole population:
we have what is called a census
any parameter of interest are deterministic
Data: ONLY a subset (or part) of the designated population, called a
sample, is available
Infer/Guess the unknown parameter (defined by all entities in the
population) based on some statistics (computed from the sample –
part of, but not all, information in the population)
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 19 / 23
Set–up of Statistical Problems II
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 20 / 23
Example 4: How Popular is Facebook?
1 Problem: Suppose that we are interested in how popular is in
Singapore
2 Population: Everyone ( , ) who resides in (i.e., experimental
units)
3 Parameter: Proportion of the Singaporean population who have a
a/c
4 Sample/data: students who are present here at UT AUD 3 in the
first lecture of ST2334 today
5 Statistic: Proportion of students who have a a/c among the
sample (all students who are present here at UT AUD 3 in the first
lecture of ST2334 today)
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 21 / 23
Why Learns Probability in Statistics Classes? I
Statistical theory is derived from mathematical theory of probability
Mastery of probability leads to a better understanding of how
statistical methodologies are developed & used
A probability problem can be described in terms of population &
sample: View a chance phenomenon as a population & outcome
from the chance phenomenon as a sample ⇔ information of the
population under study is known, & questions regarding a sample
taken from the population are posed & answered
Probability Problem
Chance phenomenon ⇔ Population
⇓
Outcome ⇔ Sample
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 22 / 23
Why Learns Probability in Statistics Classes? II
The relationship between the two disciplines can be summarized by
saying that probability reasons from the population to the
sample (deductive reasoning), whereas statistics reasons from the
sample to the population (inductive reasoning)
Before we can retrieve relevant information about the whole
population from any sample, we should first understand the
uncertainty associated with taking a sample from a given population
We always study probability before learning statistics!
Monday & Thursday, 12nn-2pm (UT AUD 3) Dr. HO Man-Wai (stahmw@[Link]) ST2334(NUS) Ch.0 23 / 23