Scribd
Upload
51 views

MAT 271 Probability and Statistics Lecture 1: Introduction: Asst. Prof. N. Kemal Ure

The document introduces a lecture on probability and statistics, explaining that probability theory provides tools for estimating chances of uncertain events and statistics allows estimating probabilities from limited data samples, and gives examples of how they are applied in fields like health, safety, manufacturing, social sciences, artificial intelligence, and forecasting.

Uploaded by

Serkan Burak ÖRS
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
51 views

MAT 271 Probability and Statistics Lecture 1: Introduction: Asst. Prof. N. Kemal Ure

The document introduces a lecture on probability and statistics, explaining that probability theory provides tools for estimating chances of uncertain events and statistics allows estimating probabilities from limited data samples, and gives examples of how they are applied in fields like health, safety, manufacturing, social sciences, artificial intelligence, and forecasting.

Uploaded by

Serkan Burak ÖRS
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 28

MAT 271 Probability and Statistics

Lecture 1: Introduction

Asst. Prof. N. Kemal Ure

Istanbul Technical University


[email protected]

February 11, 2020


Overview

1 Introduction

2 Why Probability Theory?

3 Why Statistics?

4 Applications

5 Summary
Introduction
Introduction

▶ The main focus of this lecture is to give semi-formal answers to


following questions:

∎ What is probability theory? Why do we need it?

∎ What is statistics? Why do we need it?

∎ Are they useful for engineering?

∎ What are some popular applications of these concepts?

▶ The objective is to give you a ”big picture” view of probability theory


and statistics and motivate you for the rest of the class.
Why Probability Theory?
Introduction

▶ In both social life and in engineering, we are often faced with


uncertain situations:

∎ What is the chance that I will make it to the class in time?

∎ What is the chance that my UAV will survive severe weather?

▶ Probability theory offers us the tools that take these informal


statements and turns them into mathematical objects

∎ So that we can do computation with them!

∎ So that we can estimate the probabilities of these uncertain events.

▶ That is cool, but why do we need to estimate these probabilities?


Decision Making

▶ That is cool, but why do we need to estimate these probabilities?


∎ Because understanding probability theory is the key to good
decision making.

▶ Without thinking about it consciously, you make your even daily


decisions based on probabilities
∎ Example: Selecting your transportation route to the school
∎ Example: Selecting courses from different professors
∎ Example: Betting and TV Quiz Shows

▶ Nice, but do we really need advanced mathematics to deal with these


simple problems?
Decision Trees

▶ Let’s look at a concrete example to understand why probability based


decision making can be non-trivial. (Example taken from Bertsimas’
Data, Decision and Models)

▶ Bill is a 3rd year engineering student and he needs to pick a summer


job (paid internship). He faces several decisions:

∎ He meets with the VP of a famous company (Vanessa), she tells him that
she is impressed with him and she should apply to her company.

∎ The manager at the previous company (John) he interned at offers him


12, 000 USD for the summer.

∎ He can also apply to his school’s career office, and they can find a
summer job for him.
Decision Trees

▶ How should Bill decide?

∎ He needs to reply to John as soon as possible. Vanessa’s offer is not on


the table yet. School can wait.

∎ Note that there is a sequence of decisions, not a single one.


Decision Trees

▶ First decision, accept or reject


John’s offer

▶ Next, is Vanessa going to make


an offer?:
∎ This is a probabilistic event that
we have no control over
∎ Note the differences between

circles and squares


Decision Trees

▶ Second decision, accept or reject Vanessa:

▶ Outside sources tell that her average offer is around 14, 000 USD.
Decision Trees

▶ Final link in the chain, apply school’s corporate program.

∎ But what is the average salary for that? That is where the data comes in:

∎ Based on this information, we can complete the final branch of our


decision tree
Decision Trees

▶ Here is the final branch


Decision Trees

▶ Now let’s put in the other information (and estimations) we have


Decision Trees

▶ Now let’s put in the other information (and estimations) we have


Decision Trees

▶ Next, we evaluate decision tree by multiplying gains and probabilities


Why Statistics?
Statistics

▶ Note that in the previous example, probabilities played a key role in


solving the problem.
∎ But if they are not handed to us, how do we compute those probabilities?
∎ We need to look at the data!

▶ Statistics offers us the tools for computing probabilities (and related


information) from data.
▶ Here is a picture that portrays the relationship between them (taken
from Wasserman’s All of Statistics)
Statistics

▶ Statistics is all about drawing probabilistic conclusions from limited


number of samples
Statistics

▶ Statistics is all about drawing probabilistic conclusions from limited


number of samples
An Important Remark

▶ One of the pitfalls of probability is it seems intuitive and easy, but


actually it is very deceptive and error-prone most of the time

▶ Try answering questions below intuitively

∎ A test for a rare disease (1 in a thousand) gives true positive results with
probability 0.95. If your test is positive, what is the probability that you
are infected?
∎ In a party of 10, what is the probability that two people has the same
birthday? What if there were 30 people?
∎ A cargo plane has 8 adjacent crate slots and whenever two heavy crates
are next to each other plane might become unstable. Given that there is a
7% chance of loading a heavy crate to the airplane, what is the
probability that aircraft might become unstable?
An Important Remark

▶ Same goes for statistics. It is actually very easy to lie using statistics
▶ Sample Bias:
∎ All my friends listen to metal music, therefore metal music is the most
popular genre
∎ A banking company conducts a mobile survey and the results are ’70% of
the customers are interested in using a mobile baking app’. Are these
results reliable?

▶ Correlation and Causation:


∎ Sale of ice-creams increases sales of sun glasses
∎ Sending more emails at work gets you promoted

▶ You will learn how to overcome these pitfalls in this course. After
this course ends you will know how to think and answer correctly in
probabilistic problems.
Applications
Applications

▶ Probability and statistics has tons of real-life applications. One of the


major objectives of this course is to familiarize you with them.
▶ Here are some applications we will take a look at throughout the
course

▶ Health
∎ Life expectancy
∎ Drug Trials
∎ Medical Tests

▶ Safety
∎ Traffic accidents
∎ Aviation safety
∎ Workload estimation
Applications

▶ Networks
∎ Transportation network
delays
∎ Communication network

reliability

▶ Manufacturing Processes
∎ Yield estimation
∎ Product quality control

▶ Social Sciences
∎ Education
∎ Environments
∎ Demography
Applications

▶ Artificial Intelligence
∎ Computer Vision
∎ Language Processing

▶ Autonomous Systems
∎ Perception
∎ Planning and Decision
Making

▶ Forescasting
∎ Finance
∎ Weather Prediction
∎ Retail
Summary
Summary

▶ This lecture:

∎ Why probability and statistics are important.

∎ Some popular applications of probability and statistics

▶ What is next?

∎ We start our journey into probability theory with sample spaces and
probability laws.

You might also like