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

Welcome to Scribd!

  • 36 views

    N Be A Random Sample From The Distribution

    Uploaded by

    d
    The document provides solutions to various probability problems involving a gamma distribution with probability density function f(x; λ) = 2λ2x3e-λx2 for x > 0. Part m) finds the expected value of the reciprocal of the sum of squared gamma random variables Y = ΣXi2. It states that Y has a Gamma(α = 2n, θ = 1/λ) distribution and calculates E(1/Y) = λ/(2n-1).

    Copyright:

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

    N Be A Random Sample From The Distribution

    Uploaded by

    d
    36 views13 pages
    The document provides solutions to various probability problems involving a gamma distribution with probability density function f(x; λ) = 2λ2x3e-λx2 for x > 0. Part m) finds the expected value of the reciprocal of the sum of squared gamma random variables Y = ΣXi2. It states that Y has a Gamma(α = 2n, θ = 1/λ) distribution and calculates E(1/Y) = λ/(2n-1).

    Original Title

    Estimation2ans.pdf

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document provides solutions to various probability problems involving a gamma distribution with probability density function f(x; λ) = 2λ2x3e-λx2 for x > 0. Part m) finds the expected value of the reciprocal of the sum of squared gamma random variables Y = ΣXi2. It states that Y has a Gamma(α = 2n, θ = 1/λ) distribution and calculates E(1/Y) = λ/(2n-1).

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    36 views13 pages

    N Be A Random Sample From The Distribution

    Uploaded by

    d
    The document provides solutions to various probability problems involving a gamma distribution with probability density function f(x; λ) = 2λ2x3e-λx2 for x > 0. Part m) finds the expected value of the reciprocal of the sum of squared gamma random variables Y = ΣXi2. It states that Y has a Gamma(α = 2n, θ = 1/λ) distribution and calculates E(1/Y) = λ/(2n-1).

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 13

    STAT 410

    1.

    Let > 0 and let X 1 , X 2 , , X n be a random sample from the distribution


    with the probability density function

    f ( x; ) = 2 2 x 3 e x ,
    2

    a)

    Find E ( X k ), k > 4.

    x > 0.
    Consider u = x 2 or u = x 2.

    ( a ) = u a 1 e u du , a > 0.

    Hint 1:
    Hint 2:

    E( Xk) =

    x k 2 2 x 3 e x dx
    2

    u = x2

    du = 2 x

    u 2
    =
    0
    =

    k 2

    +1

    du =

    k 2

    k
    2

    +1

    e u du

    + 2 .
    2

    OR

    E( Xk) =

    x k 2 2 x 3 e x dx
    2

    u = x2

    du = 2 x

    =
    2

    +1

    e u du

    k 2

    2
    1
    k

    + 2

    2
    0 k
    + 2

    since

    1
    k

    + 2
    2

    +2

    +1

    +2

    +1

    e u du = k 2

    + 2 ,
    2

    k
    e u is the p.d.f. of Gamma ( = + 2 , =
    2

    ).

    b)

    Obtain a method of moments estimator of , .

    That is, if E ( X ) = h ( ), solve X = h ( ) for .


    Suppose n = 5,

    x 1 = 0.6, x 2 = 1.1, x 3 = 2.7, x 4 = 3.3, x 5 = 4.5.

    and

    Find a method of moments estimate of .

    E( X ) =

    1 2

    3
    5
    1 2 3
    1 2
    1

    + 2 =

    =
    2 2
    2

    1 2

    = X

    x 1 = 0.6,
    ~
    1 =

    ( )

    ~
    1 =

    ( )2

    16 X

    x 2 = 1.1,

    16 x

    1
    1 2 3 1

    =
    2 2
    2

    3 1
    2 2

    x 3 = 2.7,

    x 4 = 3.3,

    x 5 = 4.5.

    x =

    12.2
    = 2.44.
    5

    0.09448 0.29682.

    OR

    E( X2) =

    = X

    2 2

    2
    1
    2

    .
    + 2 = ( 3 ) = 1 2 ! =

    X i2
    n
    i =1

    ~
    2 =

    2
    X2

    2n
    n

    Xi

    i =1

    x 1 = 0.6,
    ~
    2 =

    x 2 = 1.1,

    2n
    n

    x i2

    i =1

    x 3 = 2.7,

    = 0.25.

    x 4 = 3.3,

    x 5 = 4.5.

    x i2

    i =1

    = 40.

    c)

    Obtain the maximum likelihood estimator of , .


    That is, find = arg max L ( ) = arg max ln L ( ),
    where L ( ) =

    f ( xi ; ) .
    i =1

    Suppose n = 5,

    and

    x 1 = 0.6, x 2 = 1.1, x 3 = 2.7, x 4 = 3.3, x 5 = 4.5.

    Find the maximum likelihood estimate of .

    L( ) =

    2 2 x i3 e

    x i2

    i =1

    ( )

    ln L ( ) = n ln 2 + 2 n ln + ln x i3 x i2 .
    i =1

    ( ln L ( ) )' =

    n
    x i2 = 0.
    i =1

    2n

    i =1

    2n
    n

    X i2

    i =1

    x 1 = 0.6,
    =

    2n
    n

    x 2 = 1.1,

    x i2

    i =1

    = 0.25.

    x 3 = 2.7,

    x 4 = 3.3,

    x 5 = 4.5.

    x i2

    i =1

    = 40.

    d)

    What is the probability distribution of W = X 2 ?

    Let W = X 2

    fW(w ) = fX(v(w )) | v'( w) | = 2 2 w3


    =

    e)

    ( 2 )

    w 21 e

    v'( w) =

    W = v(W )

    X=

    ew

    1
    2 w

    1
    2 w

    = 2 we

    w > 0.

    W has Gamma ( = 2, =

    ) distribution.

    n
    What is the probability distribution of Y = X i2 ?
    i =1

    Suppose X and Y are independent, X is Gamma ( 1 , ), Y is Gamma ( 2 , ).


    If random variables X and Y are independent, then M X + Y ( t ) = M X ( t ) M Y ( t ).
    1

    M X + Y( t ) =

    X + Y is Gamma ( 1 + 2 , );

    i =1

    X i2 =

    i =1

    ( 1 t ) 1 ( 1 t ) 2

    ( 1 t ) 1 + 2

    W i has Gamma ( = 2 n, =

    ) distribution.

    t<

    f)

    Suppose = 0.2. Find P ( X > 5 ) = P ( X 2 > 25 ).


    Hint:

    If T has a Gamma ( , = 1 ) distribution, where is an integer, then

    F T ( t ) = P ( T t ) = P ( Y ) and P ( T > t ) = P ( Y 1 ),
    where Y has a Poisson ( t ) distribution.

    W = X 2 has Gamma ( = 2, =

    ) distribution.

    P ( X > 5 ) = P ( X 2 > 25 ) = P ( W > 25 )


    = P ( Poisson ( 0.2 25 ) 1 )
    = P ( Poisson ( 5 ) 1 )
    = 0.040.

    g)

    Suppose = 0.2. Find P ( 3 < X < 4 ).


    Hint:

    If T has a Gamma ( , = 1 ) distribution, where is an integer, then

    F T ( t ) = P ( T t ) = P ( Y ) and P ( T > t ) = P ( Y 1 ),
    where Y has a Poisson ( t ) distribution.

    W = X 2 has Gamma ( = 2, =

    ) distribution.

    P ( 3 < X < 4 ) = P ( 9 < W < 16 ) = P ( W > 9 ) P ( W > 16 )


    = P ( Poisson ( 0.2 9 ) 1 ) P ( Poisson ( 0.2 16 ) 1 )
    = P ( Poisson ( 1.8 ) 1 ) P ( Poisson ( 3.2 ) 1 )
    = 0.463 0.171 = 0.292.

    h)

    n
    Suppose n = 5 and = 0.2. Find P ( X i2 < 35 ).
    i =1
    If T has a Gamma ( , = 1 ) distribution, where is an integer, then

    Hint:

    F T ( t ) = P ( T t ) = P ( Y ) and P ( T > t ) = P ( Y 1 ),
    where Y has a Poisson ( t ) distribution.

    X i2

    has Gamma ( = 2 n, =

    i =1

    X i2

    ) distribution.

    has Gamma ( = 10, = 5 ) distribution.

    i =1

    n
    P ( X i2 < 35 ) = P ( Gamma ( = 10, = 5 ) < 35 )
    i =1
    = P ( Poisson ( 0.2 35 ) 10 ) = 1 P ( Poisson ( 7 ) 9 )
    = 1 0.830 = 0.170.

    i)

    Suppose = 0.2. Find P ( 1.10 < X < 4.41 ).


    Hint:

    If T has a Gamma ( , = 1 ) distribution, where is an integer, then


    2T
    = 2 T has a 2 ( 2 ) distribution ( a chi-square distribution with

    2 degrees of freedom ).

    W = X 2 has Gamma ( = 2, =

    ) distribution.

    P ( 1.10 < X < 4.41 ) = P ( 1.2100 < W < 19.4481 )


    = P ( 2 0.2 1.2100 < 2 ( 4 ) < 2 0.2 19.4481 )
    = P ( 0.484 < 2 ( 4 ) < 7.77924 ) = 0.90 0.025 = 0.875.

    j)

    Suppose = 0.2. Find a such that P ( X 2 < a ) = 0.10.


    Hint:

    If T has a Gamma ( , = 1 ) distribution, where is an integer, then


    2T
    = 2 T has a 2 ( 2 ) distribution ( a chi-square distribution with

    2 degrees of freedom ).

    W = X 2 has Gamma ( = 2, =

    ) distribution.

    P ( X 2 < a ) = P ( W < a ) = P ( 2 ( 4 ) < 2 0.2 a ) = 0.10.

    02.90 ( 4 )

    0.4 a =

    a = 2.66.

    2.66 1.63095.

    = 1.064.

    k)

    Suppose = 0.2. Find b such that P ( X > b ) = 0.05.


    Hint:

    If T has a Gamma ( , = 1 ) distribution, where is an integer, then


    2T
    = 2 T has a 2 ( 2 ) distribution ( a chi-square distribution with

    2 degrees of freedom ).

    W = X 2 has Gamma ( = 2, =

    ) distribution.

    P ( X > b ) = P ( X 2 > b 2 ) = P ( W > b 2 ) = P ( 2 ( 4 ) > 2 0.2 b 2 ) = 0.05.

    0.4 b 2 =

    b =

    02.05 ( 4 )

    = 9.488.

    23.72 4.87032.

    l)

    n
    Suppose n = 5 and = 0.2. Find c such that P ( X i2 < c ) = 0.01.
    i =1
    Hint:

    If T has a Gamma ( , = 1 ) distribution, where is an integer, then


    2T
    = 2 T has a 2 ( 2 ) distribution ( a chi-square distribution with

    2 degrees of freedom ).

    X i2

    has Gamma ( = 2 n, =

    i =1

    X i2

    ) distribution.

    has Gamma ( = 10, = 5 ) distribution.

    i =1

    n
    n
    P ( X i2 < c ) = P ( 2 0.2 X i2 < 2 0.2 c ) = P ( 2 ( 20 ) < 0.4 c ) = 0.01.
    i =1
    i =1

    02.99 ( 20 )

    0.4 c =

    c = 20.65.

    = 8.26.

    m)

    n
    1
    Let Y = X i2 . Find E (
    Y
    i =1

    X i2

    Y =

    E(

    Xi

    Hint: Recall part (e).

    has Gamma ( = 2 n, =

    i =1

    i =1

    Wi

    ).

    1
    = E(
    Y

    ) distribution.

    2 n 2 n1 x
    dx
    x
    e
    x ( 2n )

    i =1

    2 n1 2 n2 x

    .
    =
    x
    dx =
    e

    ( 2n 1) 0 ( 2n 1)
    2n 1

    n)

    Is the maximum likelihood estimator of , , an unbiased estimator of ?


    If is not an unbiased estimator of , construct an unbiased estimator of
    based on .

    2n
    E ( ) = E (
    Y

    2n
    .
    = +
    2n 1
    2n 1

    is NOT an unbiased estimator of .

    Consider

    2n 1
    2n 1
    .
    =
    =
    n
    2n
    2

    Xi

    i =1

    Then

    2n 1
    E ( ) =
    E ( ) = .
    2n

    o)

    Find MSE ( ) = E [ ( )

    E[(

    x2

    ( bias ( ) ) 2 + Var ( ).

    2 n 2 n1 x
    dx
    x
    e
    ( 2n )

    ( 2 n 1 )( 2 n 2 )
    2

    ( 2 n 1 )( 2 n 2 )

    1
    Var ( ) = 4 n 2 Var (
    Y

    2n 1

    bias ( ) = E ( ) =

    1 2
    ) ]=
    Y

    2 n2
    x 2 n3 e x dx
    ( 2n 2 )

    2
    2
    = 4n 2

    ( 2 n 1 )( 2 n 2 ) ( 2 n 1 ) 2

    4n 2 2

    ( 2n 1 ) 2 ( 2n 2 )

    MSE ( ) = ( bias ( ) ) 2 + Var ( ) =

    (4 n 2 + 2 n 2 ) 2

    ( 2n 1 ) 2 ( 2n 2 )

    4n 2 2
    2
    +
    ( 2n 1 ) 2 ( 2n 1 ) 2 ( 2n 2 )

    ( 2n + 2 ) 2 .
    ( 2 n 1 )( 2 n 2 )

    You might also like