Math/Stat 360-1: Probability and Statistics, Washington State University

Advertisement
Similar documents
Probability distributions

Probability Theory on Coin Toss Space

Random variables, probability distributions, binomial random variable

CONTINGENCY (CROSS- TABULATION) TABLES

Basic Probability Theory (I)

Definition and Calculus of Probability

Bayes Theorem. Bayes Theorem- Example. Evaluation of Medical Screening Procedure. Evaluation of Medical Screening Procedure

Data Modeling & Analysis Techniques. Probability & Statistics. Manfred Huber

Probability: Terminology and Examples Class 2, 18.05, Spring 2014 Jeremy Orloff and Jonathan Bloom

Chapter 4 Lecture Notes

Introduction to Probability

PROBABILITY. The theory of probabilities is simply the Science of logic quantitatively treated. C.S. PEIRCE

3.4. The Binomial Probability Distribution. Copyright Cengage Learning. All rights reserved.

ST 371 (IV): Discrete Random Variables

Chapter 4 Probability

V. RANDOM VARIABLES, PROBABILITY DISTRIBUTIONS, EXPECTED VALUE

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.436J/15.085J Fall 2008 Lecture 5 9/17/2008 RANDOM VARIABLES

An Introduction to Basic Statistics and Probability

Lesson 1. Basics of Probability. Principles of Mathematics 12: Explained! 314

Probability OPRE 6301

Summary of Formulas and Concepts. Descriptive Statistics (Ch. 1-4)

Bayesian Tutorial (Sheet Updated 20 March)

People have thought about, and defined, probability in different ways. important to note the consequences of the definition:

Math 141. Lecture 3: The Binomial Distribution. Albyn Jones 1. 1 Library jones/courses/141

Chapter What is the probability that a card chosen from an ordinary deck of 52 cards is an ace? Ans: 4/52.

Math/Stats 425 Introduction to Probability. 1. Uncertainty and the axioms of probability

STATISTICS HIGHER SECONDARY - SECOND YEAR. Untouchability is a sin Untouchability is a crime Untouchability is inhuman

Chapter 3: DISCRETE RANDOM VARIABLES AND PROBABILITY DISTRIBUTIONS

Statistics in Geophysics: Introduction and Probability Theory

Chapter 3: DISCRETE RANDOM VARIABLES AND PROBABILITY DISTRIBUTIONS. Part 3: Discrete Uniform Distribution Binomial Distribution

6. Jointly Distributed Random Variables

Sample Space and Probability

MAS108 Probability I

I. WHAT IS PROBABILITY?

3 Multiple Discrete Random Variables

SCHOOL OF ENGINEERING & BUILT ENVIRONMENT. Mathematics

ST 371 (VIII): Theory of Joint Distributions

Homework 8 Solutions

Chapter 4 - Practice Problems 1

STA 256: Statistics and Probability I

The Central Limit Theorem Part 1

MCA SEMESTER - II PROBABILITY & STATISTICS

Homework 3 Solution, due July 16

P(X = x k ) = 1 = k=1

Probability for Computer Scientists

The Calculus of Probability

Welcome to Stochastic Processes 1. Welcome to Aalborg University No. 1 of 31

Lecture 2: Introduction to belief (Bayesian) networks

Chapter 5. Discrete Probability Distributions

5. Continuous Random Variables

Probability & Probability Distributions

Chapter 9 Monté Carlo Simulation

Introduction to Probability

IAM 530 ELEMENTS OF PROBABILITY AND STATISTICS INTRODUCTION

Discrete Structures for Computer Science

Statistical Inference. Prof. Kate Calder. If the coin is fair (chance of heads = chance of tails) then

Lecture 1 Introduction Properties of Probability Methods of Enumeration Asrat Temesgen Stockholm University

Random variables P(X = 3) = P(X = 3) = 1 8, P(X = 1) = P(X = 1) = 3 8.

Machine Learning Math Essentials

For two disjoint subsets A and B of Ω, say that A and B are disjoint events. For disjoint events A and B we take an axiom P(A B) = P(A) + P(B)

Business Statistics 41000: Probability 1

Chapter 4. Probability and Probability Distributions

Unit 19: Probability Models

MAS131: Introduction to Probability and Statistics Semester 1: Introduction to Probability Lecturer: Dr D J Wilkinson

Elements of probability theory

Chapter 5. Random variables

Chapter 13 & 14 - Probability PART

Lecture Note 1 Set and Probability Theory. MIT Spring 2006 Herman Bennett

Probability and statistics; Rehearsal for pattern recognition

Chapter 3: The basic concepts of probability

Math 141. Lecture 2: More Probability! Albyn Jones 1. jones/courses/ Library 304. Albyn Jones Math 141

Probability and statistical hypothesis testing. Holger Diessel

+ Section 6.2 and 6.3

Introduction to Probability

STAT/MTHE 353: Probability II. STAT/MTHE 353: Multiple Random Variables. Review. Administrative details. Instructor: TamasLinder

A Tutorial on Probability Theory

Chapter 4 & 5 practice set. The actual exam is not multiple choice nor does it contain like questions.

M2S1 Lecture Notes. G. A. Young ayoung

Bayesian Updating with Discrete Priors Class 11, 18.05, Spring 2014 Jeremy Orloff and Jonathan Bloom

Question: What is the probability that a five-card poker hand contains a flush, that is, five cards of the same suit?

Probability and Statistics Vocabulary List (Definitions for Middle School Teachers)

INTRODUCTORY SET THEORY

Machine Learning.

Graphs. Exploratory data analysis. Graphs. Standard forms. A graph is a suitable way of representing data if:

Reliability Applications (Independence and Bayes Rule)

Master s Theory Exam Spring 2006

Chapter 4. Probability Distributions

BNG 202 Biomechanics Lab. Descriptive statistics and probability distributions I

Section 6.2 Definition of Probability

Notes on Probability. Peter J. Cameron

Section 6-5 Sample Spaces and Probability

E3: PROBABILITY AND STATISTICS lecture notes

P (A) = lim P (A) = N(A)/N,

PROBABILITY. Chapter. 0009T_c04_ qxd 06/03/03 19:53 Page 133

STAT 315: HOW TO CHOOSE A DISTRIBUTION FOR A RANDOM VARIABLE

Chapter 4 - Practice Problems 2

Pattern matching probabilities and paradoxes A new variation on Penney s coin game

A Short Introduction to Probability

Chapter 3. Probability

Definition: Suppose that two random variables, either continuous or discrete, X and Y have joint density

Advertisement
Transcription:

Math/Stat 360-1: Probability and Statistics, Washington State University Haijun Li lih@math.wsu.edu Department of Mathematics Washington State University Week 3 Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 1 / 31

Outline 1 Section 2.4: Conditional Probability 2 Section 2.5: Independence 3 Section 3.1: Random Variables 4 Section 3.2: Probability Distributions for Discrete Random Variables Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 2 / 31

Probabilistic Modeling Three Basic Ingredients: 1 Sample space Ω 2 Events E 3 Probability measure P(E) Motivation for Conditional Probability Measures It should be easier to estimate probabilities if more relevant information is given. The probability P(E) can be calculated by analyzing what could possibly happen under various possible scenarios. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 3 / 31

Cosindering three possible scenarios... Figure: If B 1 occurs, then A occurs. If B 3 occurs, then A will not occur. If B 2 occurs, likelihood of A depends on P(A B 2 ). Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 4 / 31

Definition Let A and B be two events with P(B) > 0. The conditional probability of A given that B occurs is defined as P(A B) := P(A B), A, B Ω. P(B) Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 5 / 31

Example In a city, 60% of all households get Internet service from the local cable company, 80% get television service from that company, and 50% get both services from that company. Let A = {getting Internet service}, B = {getting TV service}. 1 What is the probability that a randomly selected household gets Internet service given that it gets TV service from that company? P(A B) = P(A B) P(B) = 0.5 0.8 = 0.625. 2 What is the probability that a randomly selected household gets TV service given that it gets Internet service from that company? P(B A) = P(A B) P(A) = 0.5 0.6 = 0.833. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 6 / 31

The Multiplication Rule Theorem 1 For any two events A 1, A 2 Ω, P(A 1 A 2 ) = P(A 1 A 2 )P(A 2 ) = P(A 2 A 1 )P(A 1 ). 2 For any three events A 1, A 2, A 3 Ω, P(A 1 A }{{} 2 A 3 ) = P(A 3 A 1 A 2 )P(A }{{} 1 A 2 ) = }{{} B B B P(A 3 A 1 A 2 )P(A 2 A 1 )P(A 1 ). 3 This can be extended to multiple events. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 7 / 31

Example Four individuals have responded to a request by a blood bank for blood donations, and their blood types are unknown. Suppose only type O+ is desired and only one of the four actually has this type. If the potential donors are selected in random order for typing, what is the probability that at least three individuals must be typed to obtain the desired type? Let A = {first type is not O+}, B = {second type is not O+} P(at least three individuals are typed) = P(A B) = P(B A)P(A) = 2 3 3 4 = 0.5. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 8 / 31

Example (cont d) Four individuals have responded to a request by a blood bank for blood donations, and their blood types are unknown. Suppose only type O+ is desired and only one of the four actually has this type. What is the probability that type O+ is typed on the third donor? Let C = {third type is O+}. P(O+ is typed on the third donor) = P(C A B)P(B A)P(A) = 1 2 2 3 3 4 = 0.25. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 9 / 31

The Total Probability Law Theorem For any events A, B Ω, P(B) = P(A B) + P(A B) = P(B A)P(A) + P(B A )P(A ). Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 10 / 31

Remark Events {A 1, A 2,..., A k } constitute a partition of sample space Ω if they are mutually exclusive and k i=1 A i = Ω. For any event B, k k P(B) = P(A i B) = P(B A i )P(A i ). i=1 where P(B A i ), i = 1,..., k, are usually easier to calculate. i=1 Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 11 / 31

Example An individual has 3 different email accounts. 70% of her messages come into account #1, whereas 20% come into account #2 and the remaining 10% into account #3. Of the messages into account #1, only 1% are spam, whereas the corresponding percentages for accounts #2 and #3 are 2% and 5%, respectively. What is the probability that a randomly selected message is spam? Let A i = {message is from account # i}, i = 1, 2, 3, B = {message is spam}. It follows from the total probability law that P(B) = P(B A 1 )P(A 1 ) + P(B A 2 )P(A 2 ) + P(B A 3 )P(A 3 ) = 0.01 0.7 + 0.02 0.2 + 0.05 0.1 = 0.016. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 12 / 31

Bayes Rule Theorem Let {A 1, A 2,..., A k } be a partition of sample space Ω. For any events B Ω, P(A j B) = P(A j B) P(B) = P(B A j )P(A j ) k i=1 P(B A, j = 1,..., k. i)p(a i ) Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 13 / 31

Example An individual has 3 different email accounts. 70% of her messages come into account #1, whereas 20% come into account #2 and the remaining 10% into account #3. Of the messages into account #1, only 1% are spam, whereas the corresponding percentages for accounts #2 and #3 are 2% and 5%, respectively. What is the probability that a randomly selected message is from account #1 given that it is spam? Let A i = {message is from account # i}, i = 1, 2, 3, B = {message is spam}. It follows from Bayes rule that = P(A 1 B) = P(B A 1 )P(A 1 ) P(B A 1 )P(A 1 ) + P(B A 2 )P(A 2 ) + P(B A 3 )P(A 3 ) 0.01 0.7 0.01 0.7 + 0.02 0.2 + 0.05 0.1 = 0.007 0.016 = 0.4375. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 14 / 31

Example (Incidence of a rare disease) Only 1 in 1000 adults is afflicted with a rare disease for which a diagnostic test has been developed. The test is such that when an individual actually has the disease, a positive result will occur 99% of the time, whereas an individual without the disease will show a positive test result only 2% of the time. If a randomly selected individual is tested and the result is positive, what is the probability that the individual has the disease? Let A 1 = individual has the disease, A 2 = individual does not have the disease, and B = positive test result. P(A 1 ) = 0.001, P(A 2 ) = 0.999, P(B A 1 ) = 0.99, P(B A 2 ) = 0.02. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 15 / 31

Example (cont d) P(A 1 B) = P(A 1 B) P(B) = 0.00099 0.00099 + 0.01998 = 0.047. Figure: Path probabilities Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 16 / 31

Learning via Bayes Rule Let H be an event of interest, and E be an event representing the evidence. Figure: P(H) is updated to P(H E). Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 17 / 31

Independence Two events A and B are independent if P(A B) = P(A). The Product Form: Two events A and B are independent is equivalent to that P(A B) = P(A)P(B). Independence and mutually exclusive are different. Two mutually exclusive events are in fact highly dependent. If A and B are independent, then P(A B) = P(A)+P(B) P(A B) = P(A)+P(B) P(A)P(B). Definition Events A 1, A 2,..., A n are mutually independent if for any subset {i 1,..., i k } {1,..., n}, P(A i1 A ik ) = P(A i1 ) P(A ik ). Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 18 / 31

Example Consider a system consisting of two components #1 and #2. Assume that components work independently of one another and P(component works) = 0.9. Components #1 and #2 are connected in series, so that system works iff both #1 and #2 work. Calculate P(system works). P(system works) = P(#1)P(#2) = 0.9 2 = 0.81. Components #1 and #2 are connected in parallel, so that system works iff either #1 or #2 works. Calculate P(system works). P(system works) = P(#1) + P(#2) P(#1)P(#2) = 0.9 + 0.9 0.9 2 = 0.99. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 19 / 31

Example (cont d) Consider a system consisting of two components #1 and #2. Assume that components work independently of one another and P(component works) = 0.9. Components #1 and #2 are connected in parallel. Given that the system fails, what is the probability that component #1 fails? P(component #1 fails system works) = = 0.9 0.99 = 0.909. P(#2 works) P(system works) Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 20 / 31

Example Consider the system of 4 components. Components 1 and 2 are connected in parallel; 3 and 4 are connected in series. If components work independently of one another and P(component works) = 0.9, calculate P(system works). P(1 or 2) = 0.99, P(3 and 4) = 0.81. P(system works) = 0.99 + 0.81 (0.99)(0.81) = 0.9981. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 21 / 31

Random Variables Random variable (RV): A function defined on the sample space. Example: Toss a coin three times. Let N = number of heads in three tosses. N(TTH) = 1, N(HHH) = 3, N(HTH) = 2, N(TTT ) = 0. Example: Sample a product from an assembly line. Let T = lifelength of the item. Discrete random variable: Its values are limited to discrete points (i.e., finite or countably infinite) on the real line. Continuous random variable: It takes on continuous measurements. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 22 / 31

Example Toss a fair coin three times. The sample space = {HHH, HHT, HTH, THH, TTH, THT, HTT, TTT }. Let N denote the number of heads in three tosses. P(N = 0) = 1/8, P(N = 1) = 3/8, P(N = 2) = 3/8, P(N = 3) = 1/8. Table: Probability Masses N = x 0 1 2 3 P(N = x) 1/8 3/8 3/8 1/8 Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 23 / 31

Discrete Random Variables The distribution of a discrete random variable X is described by the probability mass function (PMF) p(x i ) = P(X = x i ), for all the possible values x i of X. Distribution of RV X: Likelihoods or relative frequencies of various values of X. Properties of PMF 1 0 p(x) 1. 2 all x s p(x) = 1. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 24 / 31

Example Consider a group of five potential blood donors, a, b, c, d, and e, of whom only a and b have type O+ blood. Five blood samples, one from each individual, will be typed in random order until an O+ individual is identified. Let RV Y = the number of typings necessary to identify an O+ individual. Then the PMF of Y is Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 25 / 31

Example (cont d) Figure: The line graph for the PMF Figure: The histogram for the PMF Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 26 / 31

Cumulative Distribution Function = Cumulative Frequency Cumulative Distribution Function (CDF) of X F(x) = P(X x) = y:y x 1 F(x) is step-wise, non-decreasing. 2 0 F(x) 1. 3 F(x) 1 as x +. p(y). Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 27 / 31

PMF vs CDF P(a X b) = y:a y b P X(y) = F(b) F(a ). Figure: P X (x) = PMF, F X (x) = CDF Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 28 / 31

Example (Five Blood Samples) Let RV Y = the number of typings necessary to identify an O+ individual. The PMF of Y is given by The CDF of Y is given by F(x) = 0 if x < 1 0.4 if 1 x < 2 0.7 if 2 x < 3 0.9 if 3 x < 4 1 if 4 x Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 29 / 31

Example (Geometric Distribution) Consider testing items coming off an assembly line one by one until a defective item (labeled F ) is found. Let X be the number of testing items necessary to find the first defective item. If P(F) = p, find the PMF and CDF of X. Let S denote a non-defective item, and so P(S) = 1 p. The PMF of X: p(k) = P(X = k) = (1 p) k 1 p, k 1. For the CDF, for any positive integer x 1, F(x) = P(X x) = k x p(k) = x (1 p) k 1 p k=1 = 1 (1 p)x p = 1 (1 p) x. p Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 30 / 31

Example (cont d): Geometric CDFs Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 3 31 / 31