Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 38 views

    Discrete Random Variables (Chapter 3) : Example 1

    Uploaded by

    Shivneet Kumar
    y =1 ∑ (1− p + p) = 1 (Property 2). y =1 1. A random variable is a numerical variable whose value depends on the outcome of an experiment. A random variable can take on discrete or continuous values. 2. Discrete random variables take on only countable values, like counting the number of heads from coin tosses. The probability distribution of a discrete random variable provides the probability that the variable equals each possible value. 3. Important discrete probability distributions include the binomial, Bernoulli, and geometric distributions. The binomial models independent trials with two outcomes, like coin tosses. The geometric models trials until the first success, like

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Discrete Random Variables (Chapter 3) : Example 1

    Uploaded by

    Shivneet Kumar
    38 views10 pages
    y =1 ∑ (1− p + p) = 1 (Property 2). y =1 1. A random variable is a numerical variable whose value depends on the outcome of an experiment. A random variable can take on discrete or continuous values. 2. Discrete random variables take on only countable values, like counting the number of heads from coin tosses. The probability distribution of a discrete random variable provides the probability that the variable equals each possible value. 3. Important discrete probability distributions include the binomial, Bernoulli, and geometric distributions. The binomial models independent trials with two outcomes, like coin tosses. The geometric models trials until the first success, like

    Original Description:

    STAT2001

    Original Title

    Lectures Chapter 3A

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    y =1 ∑ (1− p + p) = 1 (Property 2). y =1 1. A random variable is a numerical variable whose value depends on the outcome of an experiment. A random variable can take on discrete or continuous values. 2. Discrete random variables take on only countable values, like counting the number of heads from coin tosses. The probability distribution of a discrete random variable provides the probability that the variable equals each possible value. 3. Important discrete probability distributions include the binomial, Bernoulli, and geometric distributions. The binomial models independent trials with two outcomes, like coin tosses. The geometric models trials until the first success, like

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    38 views10 pages

    Discrete Random Variables (Chapter 3) : Example 1

    Uploaded by

    Shivneet Kumar
    y =1 ∑ (1− p + p) = 1 (Property 2). y =1 1. A random variable is a numerical variable whose value depends on the outcome of an experiment. A random variable can take on discrete or continuous values. 2. Discrete random variables take on only countable values, like counting the number of heads from coin tosses. The probability distribution of a discrete random variable provides the probability that the variable equals each possible value. 3. Important discrete probability distributions include the binomial, Bernoulli, and geometric distributions. The binomial models independent trials with two outcomes, like coin tosses. The geometric models trials until the first success, like

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 10

    STAT2001_CH03A Page 1 of 10

    DISCRETE RANDOM VARIABLES (Chapter 3)


    A random variable (rv) is a numerical variable whose value depends on the outcome
    of an experiment.
    Notes: A random variable must be a number; it cannot be a letter, say.
    More precisely, a random variable is a real-valued function for which the
    domain is a sample space (Definition 2.12 on page 76 in the text).
    Example 1

    A coin is tossed twice and the sequence of Hs and Ts is observed.

    (Recall that an experiment consists of two things: what is done, and what is observed.)
    Let Y be the number of Hs which come up.
    Show that Y is a random variable.
    The experiment here has 4 possible outcomes: TT, TH, HT, TT.
    Also:

    Y = 0 if the outcome is TT
    Y = 1 if the outcome is TH or HT
    Y = 2 if the outcome is HH.

    We see that Y is a numerical variable whose value depends on the outcome of an


    experiment. Therefore Y is a random variable.
    A random variable is discrete if its possible values are either
    finite or countably infinite in number (or in other words, if those values can be listed).
    For instance, Y in Ex. 1 is a discrete random variable (because {0,1,2} is a finite set).
    Notes: If the possible values of a rv Y are 0,3,6,..., then Y is discrete.
    If the possible values of Y are all the real numbers in the set (3,7),
    then Y is not discrete; in this case we say that Y is a continuous rv (see Ch 4).
    There is another type of random variable which is neither discrete nor continuous
    but a combination of both types; such a r.v. is said to have a mixed distribution.
    This may be covered later in Chapter 4 (e.g. see Example 4.19).

    STAT2001_CH03A Page 2 of 10

    The probability that a discrete random variable Y takes on a particular value y is the
    sum of the probabilities of all sample points in the sample space S that are associated
    with y.
    We write this probability P(Y = y).
    The probability distribution (pr dsn) of a discrete random variable Y is any
    information which provides P(Y = y) for each possible value y of Y.
    This information may take the form of a list, table, function (formula) or graph.
    Example 2

    Find the pr dsn of Y in Example 1.

    P(Y = 0) = P(TT) = 1/4


    P(Y = 1) = P(TH) + P(HT) = 1/4 + /14 = 1/2
    P(Y = 2) = P(HH) = 1/4.
    The above is a list. Ys probability distribution can also be presented in other ways.
    For example:

    table

    graph

    P(Y = y)

    0
    1
    2

    1/4
    1/2
    1/4

    STAT2001_CH03A Page 3 of 10

    another graph

    formula

    1/ 4, y = 0, 2
    P(Y = y ) =
    1/ 2, y = 1

    another formula

    P (Y = y ) =

    yet another formula

    P (Y = y ) =

    1
    21+( y1)

    y = 0,1,2

    1
    ,
    2(1+ | y 1|)

    y {0,1, 2}

    Notation and terminology


    It is conventional to denote rvs by upper case letters (eg, Y, X, U) and possible values
    of those rvs by the corresponding lower case letters (eg, y, x, u).
    P(Y = y) is called the probability density function (pdf) of Y,
    and is often written p(y) (or pY ( y ) or f ( y ) or fY ( y ) ).
    P(Y = y) may also be called the density function, the density,
    or the probability mass function (pmf).

    Two properties of discrete pdfs


    1.

    0 p ( y ) 1 for all y.
    (Every value of the pdf must be at least 0 and no more than 1.)

    2.

    p( y ) = 1 .
    y

    (All the values of the pdf must add up to 1.)

    STAT2001_CH03A Page 4 of 10

    Example 3

    A coin is repeatedly tossed until the first head comes up.


    Let Y be the number of tosses.
    Derive the pdf of Y, and check that it satisfies the two properties of
    discrete pdfs.

    P(Y = 1) = P(H) = 1/2


    P(Y = 2) = P(TH) = (1/2)(1/2)
    P(Y = 3) = P(TTH) = (1/2)(1/2)(1/2), etc.
    1/ 2, y = 1

    1/ 4, y = 2
    Thus Y has pdf p ( y ) =
    1/ 8, y = 3

    etc
    (Equivalently, p ( y ) = 1/ 2 y , y = 1, 2,3,... )
    We observe that Property 1 is satisfied, since 1/2, 1/4, 1/8, ... are all between 0 and 1.
    Also,

    p( y ) = 2 + 4 + 8 + = 1 . Thus Property 2 is also satisfied.


    y

    Notes: Y is a discrete rv in this example because {1,2,3,...} is


    a countably infinite set (its elements can be listed).
    A pdf uniquely defines a rv or pr dsn.
    Thus a rv cant have 2 or more different pdfs.
    Not all functions are valid pdfs.
    For example, consider p(y) = 2/3, y = 6,10.
    This function is not a pdf, since 2/3 + 2/3 1.

    STAT2001_CH03A Page 5 of 10

    Some important discrete probability distributions


    The binomial distribution
    This has to do with experiments which involve doing something several times,
    independently, and observing the number of successes.

    Example 4

    A die is rolled 7 times. Let Y be the number of 6s which come up.


    Find Ys pdf.

    P(Y = 3) = P(Three 6s and four non-6s, in any order)


    = P(6660000) + P(6606000) + ... + P(0000666)
    123

    124

    567

    <--- 'position numbers'

    (0 here stands for 1,2,3,4 or 5 coming up, ie a 'non-6')


    3

    1
    =
    6




    5 + 1 5 1 5 + + 5
    6 6 6 6 6
    6

    7 1 3 5 4
    =
    3 6 6


    1
    6

    7
    ( identical terms)
    3

    (= 0.0781) .

    7 1 5
    Similarly, P(Y = 2) =
    2 6 6

    (= 0.2344) , etc.

    7 1 y 5 7 y
    We see that p( y ) = ,
    y 6 6

    y = 0,, 7 .

    We say that Y has a binomial distribution (with parameters 7 and 1/6).


    A random variable Y has the binomial distribution with parameters n and p if its pdf is
    of the form
    n
    p ( y ) = p y (1 p ) n y ,
    y

    y = 0, , n (n = 1, 2, 3,...; 0 p 1) .

    We write Y ~ Bin(n,p) and p ( y ) = pBin ( n , p ) ( y ) .


    We call n the number of trials and p the probability of success.
    Note: Dont confuse the two uses of p here:

    p in p(y) refers to a function


    p in p y refers to a constant, etc.

    STAT2001_CH03A Page 6 of 10

    Check that the binomial pdf is proper:


    n
    n p y (1 p ) n y = ( p + (1 p ))n = 1 (Property 2).


    y =0 y
    m
    m
    Note: We have here used the binomial theorem: (a + b)m = a x b m x

    x=0 x

    (which can be proved by indiction).


    2
    2
    2
    For example, (a + b) 2 = a 0 b 20 + a1b21 + a 2 b22 = b2 + 2ab + a 2 .
    0
    1
    2

    Example 5

    A coin is going to be tossed 10 times.


    Find the probability that 3 heads come up.

    Let Y be the number of heads that come up.


    Then Y ~ Bin(10,0.5), with pdf
    10
    10 1
    p ( y ) = 0.5 y (1 0.5)10 y = 10 ,
    y 2
    y
    10 1
    So P (Y = 3) = p (3) = 10 = 0.117 .
    3 2

    y = 0, ,10 .

    Alternatively, we can make use of tables.


    On page 839 of the text we find that: P (Y 3) = 0.172

    P (Y 2) = 0.055
    Hence P (Y = 3)

    ( = p (3) )

    (= p(0) + p(1) + p(2) + p(3))


    (= p(0) + p(1) + p(2)).

    = P (Y 3) P (Y 2) = 0.172 0.055 = 0.117.

    The Bernoulli distribution (a special case of the binomial)


    A random variable Y has the Bernoulli distribution with parameter p
    if it has the binomial distribution with parameters 1 and p.
    p,
    y =1
    The pdf of Y is then p ( y ) =
    (0 p 1) .
    1 p, y = 0
    We write Y ~ Bern(p) (or Y ~ Bin(1,p)) and p ( y ) = pBern ( p ) ( y ) .
    We call p the probability of success, as before.

    STAT2001_CH03A Page 7 of 10

    Example 6

    A coin is tossed once. Let Y be the number of heads that come up.
    What is Ys distribution? Write down Ys pdf.

    Y ~ Bern(0.5), and Ys pdf is p(y) = 0.5, y = 0,1.


    Notes: We may call Y the indicator variable for the event that heads come up
    and write this rv as Y = I(A), where A = Heads come up.
    Here, I(.) denotes the standard indicator function.
    All indicator variables have a Bernoulli distribution.
    A binomial experiment consists of several Bernoulli trials.
    A binomial rv can be thought of as the sum of several Bernoulli rvs.
    Eg, we toss two coins. Let Y = number of heads that come up.
    Then Y = Y1 + Y2 , where Yi = number of heads on ith toss (must be 1 or 0)
    = I(heads on ith toss)
    1 if heads comes up on the ith toss
    =
    0 if tails comes up on the ith toss.
    Here Y1 , Y2 ~ iid Bern(1/ 2) and Y ~ Bin(2,1/ 2) .

    The geometric distribution


    Example 7

    A die is rolled repeatedly until the first 6 comes up.


    Find the pdf of Y, the number of rolls.

    P(Y = 3) = P(Two non-6s and then a 6) = P(006) = (5 / 6)2 (1/ 6) (= 0.0231).


    Similarly, P(Y = 4) = (5 / 6)3 (1/ 6) (= 0.00386), etc.
    We see that p ( y ) = (5 / 6) y1 (1/ 6),

    y = 1, 2,3,

    We say that Y has a geometric distribution (with parameter 1/6).


    A random variable Y has the geometric distribution with parameter p if its pdf is of
    the form
    p ( y ) = (1 p ) y1 p,

    y = 1, 2,3,

    (0 p 1) .

    We write Y ~ Geo(p) and p ( y ) = pGeo( p ) ( y ) .


    We call p the probability of success, as before.

    STAT2001_CH03A Page 8 of 10

    Check that the geometric pdf is proper:

    (1 p) y1 p = p (1 p) x
    y =1

    (put x = y 1)

    x= 0

    = p

    1
    = 1 (Property 2).
    1 (1 p )

    The geometric distribution can be used to model waiting times.

    Example 8

    Suppose that the probability of an engine malfunctioning during any


    1-hr period is 0.02. Find the pr that the engine will survive 2 hours.

    Let Y be the number of 1-hr periods until the first malfunction (including the 1-hr
    period in which that malfunction occurs). Then Y ~ Geo(p), where p = 0.02.
    Eg: A malfunction in the 3rd 1-hr period means that Y = 3.

    P (Y > 2) = q y1 p

    where q = 1 p = 0.98

    y =3

    = pq 2 q y12
    y =3

    = pq 2 q x

    (after putting x = y 3)

    x=0

    = pq 2

    1
    = q 2 = 0.982 = 0.9604.
    1 q

    Note: This may be calculated more simply as:


    P (Y > 2) = 1 P (Y 2)
    = 1{ p (1) + p (2)}
    = 1 ( p + qp ) = q 2 = 0.9604 ,
    or even more simply as:
    P (Y > 2) = P(Survive 2 hours)
    = P(No failures in the first 2 hours)
    = q 2 = 0.9604.

    STAT2001_CH03A Page 9 of 10

    The hypergeometric distribution


    This has to do with sampling objects from a box, without replacement, and observing
    how many have a certain characteristic.

    Example 9

    A box has 9 marbles, of which 3 are white and 6 are black.


    You randomly select 5 marbles from the bag (without replacement).
    Find the pdf of Y, the number of white marbles amongst the selected 5.

    Number the 9 marbles 1,2,...,9, with the first 3 being white and the last 6 black.
    Then the sample points may be represented by writing 12345, 12346, ..., 56789.
    Note: We don't write 13245, because this represents the same sample point as 12345.
    I.e., the distinct sample points correspond to strings of numbers in increasing order.
    9
    We see that the number of sample points is nS = .
    5
    The sample points associated with the event Y = 2 are 12456, 12457, ..., 23789,
    36
    and the number of these is n2 = .
    23
    Note: We require 2 numbers to be from 1,2,3, and the other 3 from 4,5,6,7,8,9.
    36

    n2 23
    Hence P(Y = 2) =
    =
    9
    nS

    5
    3

    6
    4
    1
    n
    Similarly, P(Y = 1) = 1 =
    9
    nS

    5

    (=

    3(20) 10
    =
    = 0.4762).
    126
    21

    (=

    3(15)
    5
    =
    = 0.3571), etc.
    126
    14

    3 6
    5 y
    y
    We see that Y has pdf p ( y ) =
    ,
    9

    5

    y = 0,1, 2, 3 .

    We say that Y has a hypergeometric distribution (with parameters 9, 3 and 5).

    STAT2001_CH03A Page 10 of 10

    A random variable Y has the hypergeometric distribution with parameters N, r and n


    if its pdf is of the form
    r

    N r

    n y
    y
    p( y ) =
    ,
    N

    n

    y = 0, , r subject to 0 n y N r

    (N = 1,2,3,...; r = 1,2,...,N;

    n = 1,2,...,N ).

    Note: The number of black balls sampled, n y, cant be less than 0


    or more than the total number of black balls in the bag, N r.
    We write Y ~ Hyp(N,r,n) and p ( y ) = pHyp( N ,r ,n ) ( y ) .
    We may call N the number of balls (parameter), r the number of white balls,
    and n the number of sampled balls.

    Example:

    There are 10 men and 15 women in a room.


    8 persons are chosen randomly to form a committee.
    Find the probability that the committee contains 6 women.

    Solution:

    Let Y = number of women on the committee.


    Then Y ~ Hyp(25,15,8)
    and so p(6) = C(15,6)C(10,2)/C(25,8) = 0.2082.

    You might also like