chapter 07rger

Upload: prayoga-akbar

Post on 08-Jul-2018

216 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/19/2019 Chapter 07rger

    1/46

    Discrete & ContinuousProbability Distributions

    Sunu Wibirama

    Basic Probability and Statistics

    Department of Electrical Engineering and Information Technology

    Faculty of Engineering, Universitas Gadjah Mada

  • 8/19/2019 Chapter 07rger

    2/46

    OUTLINE

    ! Basic Probability Distributions

    Binomial Distributions! Poisson Distributions

    ! Normal Distributions (!)

  • 8/19/2019 Chapter 07rger

    3/46

    SOME IMPORTANT THINGS

    ! Please read Walpole 8th Chapter 5 (for Binomial andPoisson) and Chapter 6 (forNormal Distribution)

    ! You can use Table A.1, A.2,and A.3 from Walpole book(p. 751)

    ! Some exercises from thesechapters (5 and 6) will beused in Midterm Exam

  • 8/19/2019 Chapter 07rger

    4/46

    Binomial Distribution

  • 8/19/2019 Chapter 07rger

    5/46

    CASE 1

    ! Suppose that 80% of the jobs submitted to adata-processing center are of a statisticalnature.

    ! Selecting a random sample of 10 submittedjobs would be analogous to tossing anunbalanced coin 10 times, with the probability

    of observing a head (drawing a statistical job)on a single trial equal to 0.80.

  • 8/19/2019 Chapter 07rger

    6/46

    CASE 2

    ! Test for impurities commonly found in drinkingwater from private wells showed that 30% of allwells in a particular country have impurity A.

    ! If 20 wells are selected at random then it would beanalogous to tossing an unbalanced coin 20 times,with the probability of observing a head (selecting

    a well with impurity A) on a single trial equal to0.30.

  • 8/19/2019 Chapter 07rger

    7/46

    BINOMIAL EXPERIMENT

    Inspection of chip(defective or non-defective)

    Inspection of public opinian(approve or not)

    ! Inspection of digging wells

    (success or failed)

    Etc….

  • 8/19/2019 Chapter 07rger

    8/46

    BERNOULLI PROCESS

    ! An experiment often consists of repeated trials,each with two possible outcome that may belabeled success and failed

    The process is called Bernoulli Process:

    ! Consists of n repeated trials

    ! The outcome can be classified as success offailed (i.e., only 2 possible outcomes)

    The probability of success (p) remains constantfrom trial to trial

    ! The repeated trials are independent

  • 8/19/2019 Chapter 07rger

    9/46

    BINOMIAL DISTRIBUTION

    ! The number X of successes in n Bernoulli trials iscalled binomial random variable

    The probability of success is denoted by p 

    ! The probability distribution of binomial randomvariable is called binomial distribution

    b(x; n, p)

  • 8/19/2019 Chapter 07rger

    10/46

    BINOMIAL DISTRIBUTION

    ! Each success has probability, p

    ! Each failure has probability, q

    q = 1 - p

    ! x = number of successes

    ! n - x = numbe of failures

    Since independent, combination probabilities aremultiplied

    b ( x ;n , p )  =n !

     x !(n  ! x )!" # $

    % & '   p  x  q n ! x 

  • 8/19/2019 Chapter 07rger

    11/46

    GRAPH OF BINOMIAL DISTRIBUTION

    b(x;n,p)

    65 87 109 121121 43 130

  • 8/19/2019 Chapter 07rger

    12/46

    BINOMIAL CUMULATIVE DISTRIBUTION

    !=

    ==

     x

     x

     pn xb pn x B

    0

    1),;(),;(

    •   As with all other distributions,the sum of all probabilities must equal unity:

  • 8/19/2019 Chapter 07rger

    13/4613

    Table A.1: Binomial Cumulative DistributionProportion, p

    Trials  Success  0.1  0.2  0.25  0.3  0.4  0.5  0.6  0.7  0.8  0.9 

    0.9000 

    0.8000 

    0.7500 

    0.7000 

    0.6000 

    0.5000 

    0.4000 

    0.3000 

    0.2000 

    0.1000 

    1  1  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000 

    2  0  0.8100  0.6400  0.5625  0.4900  0.3600  0.2500  0.1600  0.0900  0.0400  0.0100 

    2  1  0.9900  0.9600  0.9375  0.9100  0.8400  0.7500  0.6400  0.5100  0.3600  0.1900 

    2  2  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000 

    3  0  0.7290  0.5120  0.4219  0.3430  0.2160  0.1250  0.0640  0.0270  0.0080  0.0010 

    3  1  0.9720  0.8960  0.8438  0.7840  0.6480  0.5000  0.3520  0.2160  0.1040  0.0280 

    3  2  0.9990  0.9920  0.9844  0.9730  0.9360  0.8750  0.7840  0.6570  0.4880  0.2710 

    3  3  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000 

    0.6561 

    0.4096 

    0.3164 

    0.2401 

    0.1296 

    0.0625 

    0.0256 

    0.0081 

    0.0016 

    0.0001 

    4  1  0.9477  0.8192  0.7383  0.6517  0.4752  0.3125  0.1792  0.0837  0.0272  0.0037 

    4  2  0.9963  0.9728  0.9492  0.9163  0.8208  0.6875  0.5248  0.3483  0.1808  0.0523 

    0.9999 

    0.9984 

    0.9961 

    0.9919 

    0.9744 

    0.9375 

    0.8704 

    0.7599 

    0.5904 

    0.3439 

    4  4  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000  1.0000 

  • 8/19/2019 Chapter 07rger

    14/46

    BINOMIAL DISTRIBUTION

    ! Mean

    ! Variance

     µ   =  np

    !  2=  npq

  • 8/19/2019 Chapter 07rger

    15/46

    EXAMPLE

    ! Test for impurities commonly found in drinkingwater from private wells showed that 30% of allwells in a particular country have impurity A.

    ! If a random sample of 5 wells is selected from thelarge number of wells in the country, what is theprobability that:

    Exactly 3 will have impurity A?! At least 3?

    ! Fewer than 3?

  • 8/19/2019 Chapter 07rger

    16/46

    SOLUTION (Preliminary)

    Confirm that this experiment is a binomial experiment.

    ! This experiment consists of n = 5 trials, onecorresponding to each random selected well.

    ! Each trial results in an S  (the well contains impurity A)or an F  (the well does not contain impurity A).

    !  Since the total number of wells in the country is large,

    the probability of drawing a single well and finding thatit contains impurity A is equal to 0.30 and thisprobability will remain the same for each of the 5selected wells.

  • 8/19/2019 Chapter 07rger

    17/46

    SOLUTION (a)

    ! Therefore, the sampling process represents abinomial experiment with n = 5 and p = 0.30.

    a) The probability of drawing exactly x = 3 wellscontaining impurity A, with n = 5, p = 0.30 and x = 3

    b(3;5, 0.30)  =5!

    3!2!(0.30)3(1

    !0.30)5

    !

    3 = 0.1323

  • 8/19/2019 Chapter 07rger

    18/46

    SOLUTION (b)

    b) The probability of observing at least 3 wellscontaining impurity A is:

    P (x ! 3) = p(3)+p(4)+p(5).

    We have calculated p(3) = 0.1323 p(4) = 0.02835, p(5) = 0.00243.

    In result,

    P (x ! 3) = 0.1323+0.02835+0.00243 = 0.16380.

  • 8/19/2019 Chapter 07rger

    19/46

    SOLUTION (c)

    c) Probability of observing less than 3 wellscontaining impurity A is P (x

  • 8/19/2019 Chapter 07rger

    20/46

    Poisson Distribution

  • 8/19/2019 Chapter 07rger

    21/46

    POISSON DISTRIBUTION

    ! The experiment consists of counting the number x of times a particular event occurs during a givenunit of time

    The probability that an event occurs in a given unitof time is the same for all units

    ! The number of events that occur in one unit oftime is independent of the number that occur inother units.

    ! The mean number of events in each unit will bedenoted by the Greek letter ! 

  • 8/19/2019 Chapter 07rger

    22/46

  • 8/19/2019 Chapter 07rger

    23/46

    EXAMPLE

    ! Suppose that we are investigating the safety of adangerous intersection.

    Past police records indicate a mean of 5 accidentsper month at this intersection.

    ! Suppose the number of accidents is distributedaccording to a Poisson distribution. Calculate the

    probability in any month of exactly 0, 1, 2, 3 or 4accidents.

  • 8/19/2019 Chapter 07rger

    24/46

    SOLUTION

    Since the number of accidents is distributed accordingto a Poisson distribution and the mean number ofaccidents per month is 5, we have the probability of

    happening accidents in any month

    ! By this formula we can calculate:

     p(x)  =5 xe!5

     x!

    p(0) = 0.00674, p(1) = 0.03370, p(2) = 0.08425,

    p(3) = 0.14042, p(4) = 0.17552.

  • 8/19/2019 Chapter 07rger

    25/46

    POISSON CUMULATIVE DISTRIBUTION

    ! Consider the previous example. Suppose we want toknow cumulative distribution of probability less than 3accidents per month

    Than we have p(x < 3) = p(0) + p(1) + p(2)p(x < 3) = 0.00674+0.03370+0.08425= 0.12469

    Simplify your computation using Table A.2

    P(r;! ) =   p( x=0

    r

    !   x;! )

  • 8/19/2019 Chapter 07rger

    26/46

  • 8/19/2019 Chapter 07rger

    27/46

    Normal Distribution

  • 8/19/2019 Chapter 07rger

    28/46

    28

    NORMAL DISTRIBUTION

    Most important continuous probability distribution

    ! Graph called the “normal curve” (bell-shaped). Totalarea under the curve = 1

    Derived by DeMoivre and Gauss. Called the “Gaussian”distribution.

    Describes many phenomena in nature, industry andresearch

    Random variable, Xf(x)

    2 3 54 6

  • 8/19/2019 Chapter 07rger

    29/46

    0.0000

    0.0500

    0.1000

    0.1500

    0.2000

    0.2500

    0.3000

    0.3500

    0.4000

    0.4500

    -4.0 -2.0 0.0 2.0 4.0

       f   (  x   )

    Z-value

    2

    NORMAL DISTRIBUTION

    ! Definition: Density function of the normal

    random variable, X, with mean and variancesuch that:

     f  ( x) = n( x; µ ,!  ) =e

    !0.5  ( x! µ )

    !  

    "

    #$%

    &'

    2

    !     2" ,   ! (

  • 8/19/2019 Chapter 07rger

    30/46

    30

    DIFFERENT NORMAL DISTRIBUTION

    Different ________ f(x)

    0 1 32 4

    Different means

    f(x)

    0 1 32 4 5 6

    Different meansand _______

    f(x)

    0 1 32 4

  • 8/19/2019 Chapter 07rger

    31/46

    !"##$%$&' &)%*+, !"-'%"./'")&-

  • 8/19/2019 Chapter 07rger

    32/46

    32

    AREAS UNDER THE CURVE

    ! Area under the curve between x= x1 and x= x2 

    equals P(x1 < x < x2)

    dxedx xn x X  x P 

     x

     x

     x x

     x

    ! !    "#$

    %&'   (

    (

    ==

  • 8/19/2019 Chapter 07rger

    33/46

    STANDARD NORMAL DISTRIBUTION

    •  General density function:

    •  For standardized normaldistribution:

     f  ( x) = n( x; µ ,!  ) =e

    !  x

    2

    "

    #$%

    &'

    2

    2" 

     f  ( x) = n( x; µ ,!  ) =e

    !0.5  ( x! µ )

    !  

    "

    #$%

    &'

    2

    !     2" 

  • 8/19/2019 Chapter 07rger

    34/46

    STANDARD NORMAL DISTRIBUTION 

    ! "0ʼ1 2314 523367 !"#$%&'$()*+,

    ! "0 821 2 962: 4; < 2:7 102:72=7 76>?2@4: 4; A

    ! '=2:1;4=9?:B :4=923 =2:749 >2=?2C36 !  04 102:72=7 :4=923

    =2:749 >2=?2C36 "  x z 

     µ 

    !  

    "

    =

  • 8/19/2019 Chapter 07rger

    35/46

    35

    THE ORIGIN OF Z-DISTRIBUTION

    ! To enable use of z-table (Table A.3), transform to

    unit values! Mean = 0

    ! Variance =1!  

     µ "=

     X  Z 

    { }

    2

    2

    1

    2 2

    1

    2

    1

    ( )

    0.51 2

    0.5

    1( )2

    1

    2

    ( ; 0,1)

     x x

     x

     z 

     z 

     z 

     z 

     z 

     P x X x e dx

    e dz 

    n z dz  

     µ 

    !  

    ! "  

    ! "  

    #$ %#

      & '( )

    #

    < < =

    =

    =

    *

  • 8/19/2019 Chapter 07rger

    36/46

    Find area under normal curveExample: (P < -2.13) = 0.0166

  • 8/19/2019 Chapter 07rger

    37/46

    37

    EXAMPLE

    ! Given the standard normal distribution, find the area

    under the curve that lies to the LEFT of z = 1.84

    ! Solution: (use Table A.3)

    f(x)

    0

    z  0.0000  0.0100  0.0200  0.0300  0.0400  0.0500 

    1.7 

    0.9554 

    0.9564 

    0.9573 

    0.9582 

    0.9591 

    0.9599 

    1.8 0.9641 0.9649 0.9656 0.9664 0.9671 0.9678

  • 8/19/2019 Chapter 07rger

    38/46

    SOLVING CASE WITH Z-DISTRIBUTION

    ! Suppose we have a problem finding probability of acase in X  domain " p(X)

    We can use Z-Distribution to solve the problem

    ! We should transform X " Z

    ! “Finding X from Z”

  • 8/19/2019 Chapter 07rger

    39/46

    #?:7?:B !  ;=49 " 

  • 8/19/2019 Chapter 07rger

    40/46

    EXAMPLE 

  • 8/19/2019 Chapter 07rger

    41/46

    -),/'")&

  • 8/19/2019 Chapter 07rger

    42/46

    -),/'")&

  • 8/19/2019 Chapter 07rger

    43/46

  • 8/19/2019 Chapter 07rger

    44/46

    -),/'")&

  • 8/19/2019 Chapter 07rger

    45/46

    TUGAS INDIVIDU(sekaligus latihan untuk Ujian MIDTERM)

    ! D6=E2F2: 72=? 6C44F G23H436 IJ08 $7?@4:KL

    ! $M6=5?16 NOP IH2B6 ANAQ F21R1 0=R5FK

    ! $M6=5?16 NONJ IH2B6 ASNQ F21R1 0=2T5 255?76:01K

    $M6=5?16 SOAA IH2B6 AJSQ F21R1 32UV6=K

    ! $M6=5?16 SOAW IH2B6 AJXQ F21R1 10R76:01K

    ! D6=E2F2: 76:B2: 0R3?12: V2:B E6321Q C?32 7?H6=3RF2:Q029C28F2: ?3R10=21? H272 E2U2C2: +:72O Y2:B2: 3RH252:0R9F2: &+*+ 72: &"*

    ! .2021 U2F0R H6:BR9HR32:Q 82=? -6:?:

  • 8/19/2019 Chapter 07rger

    46/46

    THANK YOU

    andGOOD LUCK