+ All Categories
Home > Documents > 5.Random Variables and Expectations

5.Random Variables and Expectations

Date post: 03-Apr-2018
Category:
Upload: nurgazy-nazhimidinov
View: 223 times
Download: 0 times
Share this document with a friend

of 39

Transcript
  • 7/28/2019 5.Random Variables and Expectations

    1/39

    1

    RANDOM VARIABLES

    AND EXPECTATION

  • 7/28/2019 5.Random Variables and Expectations

    2/39

    2

    Random Variables

    A random variable is a function or rule thatassigns a numerical value to each simpleevent in a sample space.

    A random variable reflects the aspect of arandom experiment that is of interest for us.

    There are two types of random variables:

    Discrete random variable Continuous random variable.

  • 7/28/2019 5.Random Variables and Expectations

    3/39

    3

    Random Variables

    IfXis a function that assigns a realnumbered value to every possible event ina sample space of interest,Xis called a

    random variable. It is denoted by capital letters such asX, Y

    andZ.

    The specified value of the random variableis unknown until the experimental outcomeis observed.

  • 7/28/2019 5.Random Variables and Expectations

    4/39

    4

    EXAMPLES

    The experiment of flipping a fair coin.Outcome of the flip is a random variable.

    S={H,T}X(H)=1 andX(T)=0

    Select a student at random from allregistered students at METU. We want toknow the weight of these students.

    X= the weight of the selected student

    S: {x: 45kg X 300kg}

  • 7/28/2019 5.Random Variables and Expectations

    5/39

    5

    A random variable is discrete if it can assumea countable number of values.

    A random variable is continuous if it can

    assume an uncountable number of values.

    0 11/21/41/16

    Continuous random variable

    After the first value is defined

    the second value, and any valuethereafter are known.

    Therefore, the number of

    values is countable

    After the first value is defined,

    any number can be the next one

    Discrete random variable

    Therefore, the number of

    values is uncountable

    0 1 2 3 ...

    Discrete and Continuous RandomVariables

  • 7/28/2019 5.Random Variables and Expectations

    6/39

    6

    A table, formula, or graph that lists all possiblevalues a discrete random variable can assume,together with associated probabilities, is calleda discrete probability distribution.

    To calculate the probability that the randomvariable X assumes the valuex,P(X = x), add the probabilities of all the simple events for which

    X is equal tox, or Use probability calculation tools (tree diagram),

    Apply probability definitions

    Discrete Probability Distribution

  • 7/28/2019 5.Random Variables and Expectations

    7/39

    7

    If a random variable can assume valuesxi, then the following must be true:

    1)x(p.2

    xallfor1)p(x0.1

    ixall

    i

    ii

    Requirements for a DiscreteDistribution

  • 7/28/2019 5.Random Variables and Expectations

    8/39

    8

    EXAMPLE

    Consider an experiment in which a fair coin istossed 3 times.

    X= The number of heads

    Lets assign 1 for head and 0 for tail. The samplespace is

    S: {TTT,TTH,THT,HTT,THH,HTH,HHT,HHH}

    Possible values ofXis 0, 1, 2, 3. Then, the

    probability distribution ofXis

    x 0 1 2 3 Total

    p(x) 1/8 3/8 3/8 1/8 1

  • 7/28/2019 5.Random Variables and Expectations

    9/39

    9

    EXAMPLE

    An automobile service facility specializing inengine tune-ups knows that 45% of all tune-ups are done on 4-cylinder automobiles,

    40% of 6-cylinder automobiles and 15% on8-cylinder automobiles.

    LetX= The number of cylinders on the next car

    to be tuned.

  • 7/28/2019 5.Random Variables and Expectations

    10/39

    10

    Example (Contd.)

    a) What is the pmf ofX?

    b) Draw the line graph for the pmf.

    c) What is the probability that the next car tuned has at least6-cylinder? More than 6-cylinder?

    d) Draw a probability histogram.

  • 7/28/2019 5.Random Variables and Expectations

    11/39

    11

    Distribution and Relative Frequencies

    In practice, often probability distributions areestimated from relative frequencies.

    ExampleA survey reveals the following frequencies(1,000s) for the number of color TVs per household.

    Number of TVs Number of Households x p(x)0 1,218 0 1218/Total = .0121 32,379 1 .3192 37,961 2 .374

    3 19,387 3 .1914 7,714 4 .0765 2,842 5 .028Total 101,501 1.000

  • 7/28/2019 5.Random Variables and Expectations

    12/39

    12

    Determining Probability of Events

    The probability distribution can be used tocalculate the probability of different events

    Example continued

    Calculate the probability of the followingevents:

    P(The number of color TVs is 3) = P(X=3)

    =.191

    P(The number of color TVs is two or more) =P(X2)=P(X=2)+P(X=3)+P(X=4)+P(X=5)=

    .374 +.191 +.076 +.028 = .669

  • 7/28/2019 5.Random Variables and Expectations

    13/39

    13

    EXAMPLE

    Find a constant c so thatp(x) satisfies theconditions of being a pmf ofX.

    2, 1,2,3,....

    ( ) 30 ,elsewhere

    x

    c xp x

    Letp(x)=x/15, x=1,2,3,4,5 be a pmf ofX.

    FindPr(X=1 or X=2).

    FindPr(1X3).

  • 7/28/2019 5.Random Variables and Expectations

    14/39

    14

    The Cumulative DistributionFunction

    IfXis a random variable, then thecumulative distribution function (cdf),denoted byF(x) is given by

    for all real numbersx. It is a non-decreasingstep function ofx and it is right continuous.

    :

    ( ) ( ) ( )

    y y x

    F x P X x p y

  • 7/28/2019 5.Random Variables and Expectations

    15/39

    15

    The Cumulative DistributionFunction

    For any two numbers a and b, a b

    P (a Xb) = F (b)F (a-)

    where a- represents the largest possibleXvalue which is less than a.

    Ifa and b are integers,

    P (a Xb) = F (b)F(a-1)

    Taking a=b,P( X=a ) = F (a)-F (a-1)

  • 7/28/2019 5.Random Variables and Expectations

    16/39

    16

    EXAMPLE

    Let X is the number of days of sick leave takenby a randomly selected employee of a largecompany during a particular year. If the max.number of allowable sick days per year is 14,

    possible values ofXare 0,1,2,,14. WithF(0)=0.58, F(1)=0.72,F(2)=0.76,F(3)=0.81,F(4)=0.88 and F(5)=0.94,find

    P(2 X5) =

    P(X = 3) =

    P(X2) =

  • 7/28/2019 5.Random Variables and Expectations

    17/39

    17

    Describing the Population/

    Probability Distribution The probability distribution represents a

    population

    Were interested in describing the population

    by computing various parameters.

    Specifically, we calculate the population

    mean and population variance.

  • 7/28/2019 5.Random Variables and Expectations

    18/39

    18

    EXPECTED VALUES

    LetXbe a rv with pdffX(x) andg(X) be afunction ofX. Then, the expected value (orthe mean or the mathematical expectation)

    ofg(X)

    X

    x

    X

    g x f x , if X is discrete

    E g X

    g x f x dx, if X is continuous

    providing the sum or the integral exists, i.e.,

  • 7/28/2019 5.Random Variables and Expectations

    19/39

    19

    EXPECTED VALUES

    E[g(X)] is finite ifE[| g(X) |] is finite.

    Xx

    X

    g x f x < , if X is discrete

    E g X

    g x f x dx< , if X is continuous

  • 7/28/2019 5.Random Variables and Expectations

    20/39

    20

    Population Mean (Expected Value)

    Given a discrete random variable X withvaluesxi, that occur with probabilitiesp(xi),the population mean ofX is.

    ixall

    ii)x(px)X(E

  • 7/28/2019 5.Random Variables and Expectations

    21/39

    21

    Let X be a discrete random variable withpossible valuesxi that occur withprobabilitiesp(x

    i

    ), and letE(xi

    ) = . Thevariance ofX is defined by

    ixall

    i

    2

    i

    22 )x(p)x()X(E)X(V

    Population Variance

    2

    isdev iationdardtansThe

  • 7/28/2019 5.Random Variables and Expectations

    22/39

    22

    EXAMPLE

    The pmf for the number of defective itemsin a lot is as follows

    0.35, 0

    0.39, 1

    ( ) 0.19, 2

    0.06, 3

    0.01, 4

    x

    x

    p x x

    x

    x

    Find the expected number and the variance of

    defective items.

  • 7/28/2019 5.Random Variables and Expectations

    23/39

    23

    EXAMPLE

    What is the mathematical expectation if wewin $10 when a die comes up 1 or 6 andlose $5 when it comes up 2, 3, 4 and 5?

    X= amount of profit

  • 7/28/2019 5.Random Variables and Expectations

    24/39

    24

    EXAMPLE

    A grab-bay contains 6 packages worth $2each, 11 packages worth $3, and 8packages worth $4 each. Is it reasonable

    to pay $3.5 for the option of selecting oneof these packages at random?

    X= worth of packages

  • 7/28/2019 5.Random Variables and Expectations

    25/39

    25

    Laws of Expected Value

    LetXbe a rv and a, b, and c be constants.Then, for any two functionsg1(x) andg2(x)whose expectations exist,

    1 2 1 2)a E ag X bg X c aE g X bE g X c

    1 10 , 0.b) If g x for all x then E g X

    1 2 1 2) .c If g x g x for all x, then E g x E g x

    1 1)d If a g x b for all x, then a E g X b

  • 7/28/2019 5.Random Variables and Expectations

    26/39

    26

    Laws of Expected

    Value

    E(c) = c

    E(X + c) = E(X) + c

    E(cX) = cE(X)

    Laws of

    Variance

    V(c) = 0

    V(X + c) = V(X)

    V(cX) = c2V(X)

    Laws of Expected Value and Variance

  • 7/28/2019 5.Random Variables and Expectations

    27/39

    27

    SOME MATHEMATICAL

    EXPECTATIONS

    Population Mean: = E(X)

    Population Variance:

    2 22 2

    0Var X E X E X (measure of the deviation from the population mean)

    Population Standard Deviation: 2 0

    Moments:* k

    k E X the k-th moment

    k

    k E X the k-th central moment

  • 7/28/2019 5.Random Variables and Expectations

    28/39

    28

    SKEWNESS

    Measure of lack of symmetry in the pdf.

    3

    3

    3 3/2

    2

    E XSkewness

    If the distribution ofXis symmetric around itsmean ,

    3=0 Skewness=0

  • 7/28/2019 5.Random Variables and Expectations

    29/39

    29

    KURTOSIS

    Measure of the peakedness of the pdf. Describesthe shape of the r,v.

    4

    4

    4 2

    2

    E X

    Kurtosis

    Kurtosis=3 Normal

    Kurtosis >3 Leptokurtic

    (peaked and fat tails)

    Kurtosis

  • 7/28/2019 5.Random Variables and Expectations

    30/39

    30

    EXAMPLE

    An appliance dealer sells 3 different modelsof upright freezers having 13.5, 15.9 and 19.1cubic feet of storage space, respectively.

    X= the amount of storage space purchased bythe next customer.

    The pmf ofX0.2, 13.5

    ( ) 0.5, 15.9

    0.3, 19.1

    x

    p x x

    x

  • 7/28/2019 5.Random Variables and Expectations

    31/39

    31

    EXAMPLE (contd.)

    a) E(X)=?

    b) If the price of a freezer having capacityXcubic feet is25X-8.5, what is the expected price paid by the next

    customer to buy a freezer?

    c) Var (X)=?

    d) Suppose that although the rated capacity of a freezer isX,the actual capacity is h (X)=X-0.01X2. What is the expectedactual capacity of the freezer purchased by the nextcustomer?

  • 7/28/2019 5.Random Variables and Expectations

    32/39

    32

    A continuous random variable has anuncountably infinite number of values inthe interval (a,b).

    Continuous Probability Distributions

    The probability that a continuous variableX will assume any particular value is zero.Why?

    0 11/21/3 2/3

    1/2 + 1/2 = 11/3 + 1/3 + 1/3 = 1

    1/4 + 1/4 + 1/4 + 1/4 = 1The probability of each value

  • 7/28/2019 5.Random Variables and Expectations

    33/39

    33

    0 11/21/3 2/3

    1/2 + 1/2 = 11/3 + 1/3 + 1/3 = 1

    1/4 + 1/4 + 1/4 + 1/4 = 1

    As the number of values increases the probability of each

    value decreases. This is so because the sum of all the

    probabilities remains 1.

    When the number of values approaches infinity (becauseX

    is continuous) the probability of each value approaches 0.

    The probability of each value

    Continuous Probability

    Distributions

  • 7/28/2019 5.Random Variables and Expectations

    34/39

    34

    To calculate probabilities we define aprobability density functionf(x).

    The density function satisfies the following

    conditionsf(x) is non-negative,

    The total area under the curve representing f(x)equals 1.

    x1 x2

    Area = 1P(x1

  • 7/28/2019 5.Random Variables and Expectations

    35/39

    35

    CUMULATIVE DISTRIBUTIONFUNCTION

    min

    ( ) ( ) ( )x

    X

    F x P X x f x dx

    Example: LetXis the amount of time for which abook on two-hour reserve at a college library is

    checked out by a randomly selected student.Suppose thatXhas density function

    0.5 ,0 2( )

    0 ,otherwise

    x xf x

    P(X1)=?

    P(0.5 X1.5)=?P(X

  • 7/28/2019 5.Random Variables and Expectations

    36/39

    36

    EXPECTED VALUE

    The expected value or mean value of acontinuous random variableXwith pdff(x) is

    ( ) ( )

    all x

    E X xf x dx

    The variance of a continuous randomvariableXwith pdff(x) is

    2 2 2

    all x

    2 2 2 2

    all x

    ( ) ( ) ( ) ( )

    ( ) ( ) ( )

    Var X E X x f x dx

    E X x f x dx

  • 7/28/2019 5.Random Variables and Expectations

    37/39

    37

    EXAMPLE

    LetXbe a random variable and it is a lifelength of light bulb. Its pdf is

    f(x)=2(1-x), 0< x < 1

    FindE(X) and Var(X).

  • 7/28/2019 5.Random Variables and Expectations

    38/39

    38

    EXPECTED VALUE OF SUMS OF

    RANDOM VARIABLES

    IfXand Yare random variables, then

    E(X+Y) = E(X) + E(Y)

  • 7/28/2019 5.Random Variables and Expectations

    39/39

    Example At computer store, the annual demand for a particular software

    package is a discrete random variable X. The store owner ordersfour copies of the package at 10$ per copy and charges customers$35 per copy. At the end of the year, the package is obsolete andthe owner loses the investment on unsold copies. The pdf ofXisgiven by the following table:

    FindE(X) and Var(X).

    Express the owners net profit Yas a linear function ofX, and findE(Y) and Var(Y).

    39

    x 0 1 2 3 4

    P(x) .1 .3 .3 .2 .1


Recommended