chapter1.1- probability theory

Upload: mohamed2002

Post on 10-Apr-2018

244 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/8/2019 Chapter1.1- Probability Theory

    1/51

    1

    OBS 615:

    MANAGERIAL DECISION

    MAKING PROCESSES ANDTECHNIQUES

  • 8/8/2019 Chapter1.1- Probability Theory

    2/51

    2

    This course is intended to enable

    students:

    Understand managerial decision-makingprocesses in organizations

    Appreciate the use of various quantitativetechniques in making decisions

    Apply the processes and techniques in real

    managerial problem solving situations

  • 8/8/2019 Chapter1.1- Probability Theory

    3/51

    3

    The following topics will be

    covered

    Quantitative methods for Decision making

    Linear programming Transportation Problems Assignment problems Network analysis

    Inventory management Waiting line methods Simulation Models

  • 8/8/2019 Chapter1.1- Probability Theory

    4/51

    4

    Chapter 1.1

    Probability

  • 8/8/2019 Chapter1.1- Probability Theory

    5/51

    5

    Learning Objectives

    Concepts: Events, Sample Space,Intersection & Union

    Tools: Venn Diagram, Tree Diagram,

    Contingency TableProbability, Conditional Probability,

    Addition Rule & Multiplication Rule

    Use Bayes TheoremSummarize the Rules ofPermutation andCombination

  • 8/8/2019 Chapter1.1- Probability Theory

    6/51

    6

    Basic Concepts

    Random Experiment is a process leadingto at least two possible outcomes withuncertainty as to which will occur.

    yA coin is thrown

    yA consumer is asked which of two

    products he or she prefersyThe daily change in an index of stock

    market prices is observed

  • 8/8/2019 Chapter1.1- Probability Theory

    7/51

    7

    Sample Spaces

    Collection ofall possible outcomes

    e.g.: All six faces of a die:

    e.g.: All 52 cards

    a deck ofbridge cards

  • 8/8/2019 Chapter1.1- Probability Theory

    8/51

    8

    Events and Sample Spaces

    An event is a set of basic outcomes fromthe sample space, and it is said to occurif the random experiment gives rise to

    one ofits constituent basic outcomes.Simple EventOutcome With 1 Characteristic

    Joint Event2 Events Occurring Simultaneously

    Compound EventOne or Another Event Occurring

  • 8/8/2019 Chapter1.1- Probability Theory

    9/51

    9

    Simple Event

    A: Male

    B: Over age 20

    C: Has 3 credit cards

    D: Red card from a deck of bridge cards

    E: Ace card from a deck of bridge cards

  • 8/8/2019 Chapter1.1- Probability Theory

    10/51

    10

    Joint Event

    D and E, (DE):

    Red, ace card from a bridge deck

    A and B, (AB):

    Male, over age 20

    among a group ofsurvey respondents

  • 8/8/2019 Chapter1.1- Probability Theory

    11/51

    11

    Intersection

    Let A and B be two events in the samplespaceS. Theirintersection, denoted AB,is the set of all basic outcomes inSthat

    belong to both A and B. Hence, the intersection AB occurs if and

    only ifboth A and B occur.

    If the events A and B have no commonbasic outcomes, their intersection AB issaid to be the empty set.

  • 8/8/2019 Chapter1.1- Probability Theory

    12/51

    12

    Compound Event

    D or E, (DE):Ace or Red card from bridge deck

  • 8/8/2019 Chapter1.1- Probability Theory

    13/51

    13

    Union

    Let A and B be two events in thesample spaceS. Theirunion, denoted

    AB, is the set of all basic outcomesinSthat belong to at least one of thesetwo events.

    Hence, the union AB occurs if andonly ifeither A or B or both occurs

  • 8/8/2019 Chapter1.1- Probability Theory

    14/51

    14

    Event Properties

    Mutually Exclusive

    Two outcomes that cannotoccur at the same time

    E.g. flip a coin, resulting inhead and tail

    Collectively ExhaustiveOne outcome in sample space

    must occur

    E.g. Male or Female

  • 8/8/2019 Chapter1.1- Probability Theory

    15/51

    15

    ii

    Special Events

    Null Event

    C

    lub & Diamond on1 Card Draw

    Complement of Event

    For Event A, AllEvents Not In A:

    A' or A

    Null Event

  • 8/8/2019 Chapter1.1- Probability Theory

    16/51

    16

    What is Probability?1.Numerical measure of

    likelihood that the event

    will occurSimple Event

    Joint Event

    Compound

    2.Lies between 0 & 1

    3.Sum of events is 1

    1

    .5

    0

    Certain

    Impossible

  • 8/8/2019 Chapter1.1- Probability Theory

    17/51

    17

    Concept of Probability

    A Prioriclassical probability, the probability ofsuccess is based on prior knowledge of the processinvolved.

    i.e. the chance of picking a black card from a

    deck of bridge cardsEmpirical classical probability, the outcomes are

    based on observed data, not on prior knowledge of aprocess.

    i.e. the chance that individual selected at randomfrom the Kalosha employee survey if satisfiedwith his or her job. (.89)

  • 8/8/2019 Chapter1.1- Probability Theory

    18/51

    18

    Concept of ProbabilitySubjective probability, the chance of

    occurrence assigned to an event by a

    particular individual, based on his/herexperience, personal opinion andanalysis of a particular situation.

    i.e. The chance of a newly designed style ofmobile phone will be successful in market.

  • 8/8/2019 Chapter1.1- Probability Theory

    19/51

    19

    (There are 2 ways to get one 6 and the other 4)

    e.g. P( ) = 2/36

    Computing Probabilities

    The probability of an event E:

    Each of the outcomes in the sample space isequally likely to occur

    number of event outcomes( )

    total number of possible outcomes in the sample space

    P E

    X

    T

    !

    !

  • 8/8/2019 Chapter1.1- Probability Theory

    20/51

    20

    Concept of Probability

    Experi- Number Number Relativementer of Trials of heads Frequency

    Demogen 2048 1061 0.

    5

    181Buffon 4040 2048 0.5069

    Pearson 12000 6019 0.5016

    Pearson 24000 12012 0.5005

    What conclusion can be drawn from the

    observations?

  • 8/8/2019 Chapter1.1- Probability Theory

    21/51

    21

    Presenting Probability &

    Sample Space

    1. Listing

    S = {Head, Tail}2. Venn Diagram

    3. Tree Diagram

    4. Contingency

    Table

  • 8/8/2019 Chapter1.1- Probability Theory

    22/51

    22

    S

    A

    Venn Diagram

    Example: Kalosha Employee Survey

    Event:A = Satisfied, = Dissatisfied

    P(A) = 356/400 = .89, P() = 44/400 = .11

  • 8/8/2019 Chapter1.1- Probability Theory

    23/51

    23

    KaloshaEmployee

    Tree Diagram

    Satisfied

    Not

    Satisfied

    Advanced

    Not

    Advanced

    Advanced

    Not

    Advanced

    P(A)=.89

    P()=.11

    .485

    .405

    .035

    .075

    Example: Kalosha Employee Survey JointProbability

  • 8/8/2019 Chapter1.1- Probability Theory

    24/51

    24

    Event

    Event B1 B2 Total

    A1 P(A1B1) P(A1B2) P(A1)

    A2 P(A2B1) P(A2B2) P(A2)

    Total P(B1) P(B2) 1

    Joint Probability

    Using Contingency Table

    Joint Probability Marginal (Simple)Probability

  • 8/8/2019 Chapter1.1- Probability Theory

    25/51

    25

    Joint ProbabilityUsing Contingency Table

    Kalosha Employee Survey

    Satisfied NotSatisfied Total

    Advanced

    NotAdvanced

    Total

    .485 .035

    .405 .075

    .52

    .48

    .89 .11 1.00

    Joint Probability

    Simple Probability

  • 8/8/2019 Chapter1.1- Probability Theory

    26/51

    26

    Use of Venn Diagram

    Fig. 3.1: AB, Intersection of events A & B,

    mutually exclusive

    Fig.3

    .2: A

    B, Union of events A & BFig. 3.3: , Complement of event A

    Fig. 3.4 and 3.5:

    The events AB and B are mutuallyexclusive, and their union is B.

    (A B) ( B) = B

  • 8/8/2019 Chapter1.1- Probability Theory

    27/51

    27

    Use ofVenn Diagram

    Let E1, E2,, Ekbe Kmutually exclusive and

    collective exhaustive events, and let A be

    some other event. Then the Kevents E1A,E2A, , EkA are mutually exclusive,

    and their union is A.

    (E1A) (E2A) (EkA) = A

    (See supplement Fig. 3.7)

  • 8/8/2019 Chapter1.1- Probability Theory

    28/51

    28

    Compound Probability

    Addition Rule1. Used to Get Compound Probabilities for

    Union of Events

    2. P(A or B) = P(A B)= P(A) + P(B) P(A B)

    3. ForMutually Exclusive Events:

    P(A or B) = P(A B) = P(A) + P(B)4. Probability ofComplement

    P(A) + P() = 1. So, P() = 1 P(A)

  • 8/8/2019 Chapter1.1- Probability Theory

    29/51

    29

    Addition Rule: Example

    A hamburger chain found that 75%

    of all customersuse mustard, 80% use ketchup, 65% use both.What is the probability that a particular customerwill use at least one of these?

    A = Customers use mustardB = Customers use ketchup

    AB = a particular customer will use at least oneof these

    Given P(A) = .75, P(B) = .80, and P(AB) = .65,P(AB) = P(A) + P(B) P(AB)

    = .75 + .80 .65= .90

  • 8/8/2019 Chapter1.1- Probability Theory

    30/51

    30

    Conditional Probability

    1. Event Probability Given that AnotherEvent Occurred

    2. Revise Original Sample Space toAccount forNew Information

    Eliminates Certain Outcomes

    3. P(A | B) = P(A and B) , P(B)>0P(B)

  • 8/8/2019 Chapter1.1- Probability Theory

    31/51

    31

    Example

    Recall the previous hamburger chainexample, what is the probability that aketchup user uses mustard?

    P(A|B) = P(AB)/P(B)

    = .65/.80 = .8125

    Please pay attention to the differencefrom the joint event in wording of thequestion.

  • 8/8/2019 Chapter1.1- Probability Theory

    32/51

    32

    S

    Black

    Ace

    Conditional Probability

    Black Happens: Eliminates All Other Outcomes andThus Increase the Conditional Probability

    Event (Ace and Black)

    (S)Black

    Draw a card, what is the probability of black ace?What is the probability of black ace when black happens?

  • 8/8/2019 Chapter1.1- Probability Theory

    33/51

    33

    ColorType Red Black Total

    Ace 2 2 4

    Non-Ace 24 24 48

    Total 26 26 52

    Conditional Probability Using

    Contingency TableConditional Event: Draw 1 Card. Note BlackAce

    RevisedSampleSpace

    P(Ace|Black) =P(Ace and Black)

    P(Black)= = 2/262/52

    26/52

  • 8/8/2019 Chapter1.1- Probability Theory

    34/51

    34

    Game Show

    Imagine you have been selected for a gameshow that offers the chance to win anexpensive new car. The car sits behind one ofthree doors, while monkeys reside behind theother two.

    You choose a door, and the host opens one ofthe remaining two, revealing a monkey.

    The host then offers you a choice: stick with

    your initial choice or switch to the other, still-unopened door.

    Do you think thatswitch to the other doorwould increase your chance of winning the car?

  • 8/8/2019 Chapter1.1- Probability Theory

    35/51

    35

    Statistical Independence

    1. Event Occurrence Does NotAffectProbability of Another Event

    e.g. Toss 1C

    oin Twice, Throw3

    Dice2. Causality Not Implied

    3. Tests For Independence

    P(A | B) = P(A), or P(B | A) = P(B),or P(A and B) = P(A)P(B)

  • 8/8/2019 Chapter1.1- Probability Theory

    36/51

    36

    Statistical Independence

    Kalosha Employee Survey

    Satisfied NotSatisfied Total

    Advanced

    NotAdvanced

    Total

    .485 .035

    .405 .075

    .52

    .48

    .89 .11 1.00

    P(A1)

    Note: (.52)(.89) = .4628{ .485

    P(B1

    )P(A1

    and B1

    )

  • 8/8/2019 Chapter1.1- Probability Theory

    37/51

    37

    Multiplication Rule

    1. Used to Get Joint Probabilities forIntersection of Events (Joint Events)

    2. P(A and B) = P(A B)P(A B) = P(A)P(B|A)

    = P(B)P(A|B)

    3. ForIndependent Events:

    P(A and B) = P(AB) = P(A)P(B)

  • 8/8/2019 Chapter1.1- Probability Theory

    38/51

    38

    Randomized Response

    An approach for solicitation ofhonest answersto sensitive questions in surveys.

    A survey was carried out in spring 1997 to

    about 150 undergraduate students takingStatistics for Business and Economics atPeking University.

    Each student was faced with two questions.

    Students were asked first to flip a coin andthen to answer question a) if the result wasthe national emblem and b) otherwise.

  • 8/8/2019 Chapter1.1- Probability Theory

    39/51

    39

    Practice of Randomized Response

    Questiona) Is the second last digit of your office phone

    number odd?

    b) Recall your undergraduate course work, have youever cheated in midterm or final exams?

    Respondents, please do as follows:

    1. Flip a coin.

    2. If the result is the national emblem (

    ), answerquestion a); otherwise answer question b).Please circleYes or No below as your answer.

    Yes No

  • 8/8/2019 Chapter1.1- Probability Theory

    40/51

    40

    Bayes Theorem

    1. Permits RevisingOldProbabilities Based on

    New Information2. Application ofConditional Probability

    3. Mutually ExclusiveEvents

    New

    Information

    Revised

    Probability

    Apply Bayes'

    Theorem

    Prior

    Probability

  • 8/8/2019 Chapter1.1- Probability Theory

    41/51

  • 8/8/2019 Chapter1.1- Probability Theory

    42/51

    42

    Bayess Theorem Example

    Fifty percent of borrowers repaid their loans. Out ofthose who repaid, 40% had a college degree. Tenpercent of those who defaulted had a college degree.What is the probability that a randomly selectedborrower who has a college degree will repay the loan?

    B1= repay, B2= default, A=college degree

    P(B1) = .5, P(A|B1) = .4, P(A|B2) = .1, P(B1|A) =?

    )()|()()|(

    )()|()|(

    2211

    111BPBAPBPBAP

    BPBAP

    ABP

    !

    8.25.

    2.

    )5)(.1(.)5)(.4(.

    )5)(.4(.!!

    !

  • 8/8/2019 Chapter1.1- Probability Theory

    43/51

    43

    Event PriorProb

    Cond.Prob

    JointProb

    Post.Prob

    Bi P(Bi) P(A|Bi) P(Bi A) P(Bi |A)

    B1 .5 .4 .20 .20/.25 = .8

    B2 .5 .1 .05 .05/.25 = .2

    1.0 P(A) = 0.25 1.0

    Bayes Theorem Example

    Table Solution

    DefaultRepay P(College)

    X =

  • 8/8/2019 Chapter1.1- Probability Theory

    44/51

    44

    Random Testing for AIDS?

    In 1987, the US Secretary of Health proposedto test blood to estimate how many Americans,and which ones, were infected with AIDS.Continuing calls have been made for

    mandatory testing to identify persons with thedisease. These calls met with considerableresistance from statisticians, who speak out

    against the mindless proposals for mandatoryAIDS testing....

    Refer to Random Testing for AIDS? Chance,

    vol. 1, no. 1, 1988

  • 8/8/2019 Chapter1.1- Probability Theory

    45/51

    45

    Permutation and Combination

    Counting Rule 1

    ExampleIn TV Series: Kangxi Empire,Shilang tossed 50 coins, the number of

    outcomes is 22 2 250

    . What is theprobability of all coins with heads up?

    If any one ofn different mutually exclusive

    and collectively exhaustive events can occuron each ofrtrials, the number of possible

    outcomes is equal tonnn nr

  • 8/8/2019 Chapter1.1- Probability Theory

    46/51

    46

    Permutation and Combination

    Application: Lottery Post Card (Post ofChina)Six digits in each group (2000 version)

    Winning the first prize: No.035718

    Winning the 5th prize: ending with number3

    If there are k1

    events on the first trial, k2

    events

    on the second trial, and krevents on the rth

    trial, then the number of possible outcomes

    is: k1k

    2k

    r

  • 8/8/2019 Chapter1.1- Probability Theory

    47/51

    47

    Permutation and Combination

    Applicatin: If a license plate consists of3 letters

    followed by 3 digits, the total number of

    outcomes would be? (most states in the US)

    Application: China License PlatesHow many licenses can be issued?

    Style 1992: one letter or digit plus 4 digits.

    Style 2002: 1) three letters + three digits2) three digits + three letters

    3) three digits + three digits

  • 8/8/2019 Chapter1.1- Probability Theory

    48/51

    48

    Permutation and Combination

    Counting Rule 2

    Example: The number of ways that 5 books could

    be arranged on a shelf is: (5)(4)(3)(2)(1) = 120

    The number of ways that all n objects can bearranged in orderis

    =n

    (n

    -1)(n

    -2)~

    (2)(1) =n!

    Where n! is called factorial and 0! isdefined as 1.

    n

    n

    P

  • 8/8/2019 Chapter1.1- Probability Theory

    49/51

    49

    Permutation and Combination

    Counting Rule 3: Permutation

    ExampleWhat is the number of ways of

    arranging 3books selected from 5books in

    order? (5)(4)(3) = 60

    The number of ways of arranging robjectsselected from nobjects in order is

    )!(!

    rn

    nP

    n

    r

    !

  • 8/8/2019 Chapter1.1- Probability Theory

    50/51

    50

    Permutation and Combination

    Counting Rule 4: CombinationExampleThe number ofcombinations of3 books

    selected from 5 books is

    (5)(4)(3)/[(3)(2)(1)] = 10

    Note: 3! possible arrangements in order are irrelevant

    The number of ways that arranging robjects selectedfrom nobjects, irrespective of the order, is equal to

    )!(!

    !

    rnr

    n

    r

    nC

    n

    r

    !

    !

  • 8/8/2019 Chapter1.1- Probability Theory

    51/51

    51

    Examples

    Applications: NBA Draft (2002)Randomly choose 4 numbers from 14 numbers

    Example:1. World Cup 2002: how many games should a

    soccer team play in a group of four teams?

    2. There are 10 cities, how many different non-stop tickets? How many different prices ofthe tickets?

    NBA Lottery.lnk