Scribd Logo
Upload

Welcome to Scribd!

  • 60 views

    Conditional Distributions and Transformations: Homework 9

    Uploaded by

    Zihan Zhu
    This document contains problems related to conditional distributions and transformations. Problem 1 involves finding the mass function of the sum of two uniformly distributed random variables. Problem 2 involves finding densities related to normal random variables, including the density of Z1^2, the density of Z1^2 + Z2^2, and showing Z1 + Z2 and Z1 - Z2 are independent. Problem 3 involves simulating pairs of standard normal random variables with various correlations using the Box-Muller transform. The challenging problems involve finding densities and distributions related to gamma and Poisson random variables.

    Copyright:

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

    Conditional Distributions and Transformations: Homework 9

    Uploaded by

    Zihan Zhu
    60 views6 pages
    This document contains problems related to conditional distributions and transformations. Problem 1 involves finding the mass function of the sum of two uniformly distributed random variables. Problem 2 involves finding densities related to normal random variables, including the density of Z1^2, the density of Z1^2 + Z2^2, and showing Z1 + Z2 and Z1 - Z2 are independent. Problem 3 involves simulating pairs of standard normal random variables with various correlations using the Box-Muller transform. The challenging problems involve finding densities and distributions related to gamma and Poisson random variables.

    Original Title

    Theory_of_Probability_HW09.pdf

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document contains problems related to conditional distributions and transformations. Problem 1 involves finding the mass function of the sum of two uniformly distributed random variables. Problem 2 involves finding densities related to normal random variables, including the density of Z1^2, the density of Z1^2 + Z2^2, and showing Z1 + Z2 and Z1 - Z2 are independent. Problem 3 involves simulating pairs of standard normal random variables with various correlations using the Box-Muller transform. The challenging problems involve finding densities and distributions related to gamma and Poisson random variables.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    60 views6 pages

    Conditional Distributions and Transformations: Homework 9

    Uploaded by

    Zihan Zhu
    This document contains problems related to conditional distributions and transformations. Problem 1 involves finding the mass function of the sum of two uniformly distributed random variables. Problem 2 involves finding densities related to normal random variables, including the density of Z1^2, the density of Z1^2 + Z2^2, and showing Z1 + Z2 and Z1 - Z2 are independent. Problem 3 involves simulating pairs of standard normal random variables with various correlations using the Box-Muller transform. The challenging problems involve finding densities and distributions related to gamma and Poisson random variables.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 6

    Conditional Distributions and Transformations

    Homework 9

    Problems
    1. Let X1 and X2 be uniform on the consecutive integers −n, −(n + 1), . . . , n − 1, n. Use convolution to
    fins the mass function for X1 + X2 .
    Let Z = X1 + X2 ,
    P (Z = z) = P (X1 + X2 = z)
    Xn
    = P (X1 = i)P (X2 = z − i)
    i=z−n
    Xn
    1 1 (1)
    =
    i=z−n
    2n + 1 2n +1
    2n + 1 − z
    =
    (2n + 1)2

    2. Let Z1 and Z2 be N (0, 1) random variables.


    (a) Find the density of Z12 .
    Suppose Z1 and Z2 are independent. Let Y1 = Z12 ,

    P (Y1 ≤ y) = P (Z12 ≤ y)
    p
    = P (Z12 ≤ (y)) (2)
    √
    = FZ1 ( y)
    √
    fY1 (y) = FZ0 1 ( y)
    √ 1
    = fZ1 ( y) √
    2 y (3)
    1 y
    = √ exp(− )
    2 πy 2

    (b) Find the density of Z12 + Z22 .

    1
    Let Y2 = Z22 . Y1 and Y2 have the same distribution. Z12 + Z22 = Y1 + Y2 ,
    Z y
    P (Y1 + Y2 = y) = P (Y1 = s)P (Y2 = y − s)ds
    0
    Z y
    1 s 1 y−s
    = √ exp(− ) p exp(− )ds (4)
    0 2 πs 2 2 π(y − s) 2
    Z y
    1 y 1
    = exp(− ) p ds
    4π 2 0 s(y − s)
    Z y Z y
    1 1
    p ds = q ds
    0 s(y − s) 0 y 2 y 2
    4 − (s − 2 )
    Z y2
    1
    = dz
    (5)
    q
    y y 2
    −2 − z 2
    4
    2s y2
    = arcsin( )| y
    y −2
    = 2π
    1 y
    So f (y) = 2 exp(− 2 ).
    (c) Show that Z1 + Z2 and Z1 − Z2 are independent and describe their distribution.

    Cov(Z1 + Z2 , Z1 − Z2 ) = Cov(Z1 , Z1 ) − Cov(Z2 , Z2 ) = 1 − 1 = 0 (6)


    So Z1 + Z2 and Z1 − Z2 are independent. They both follow N (0, 2).
    3. Use the Box Muller transform and the probability transform to simulate 100 pairs of standard normal
    random variables with correlations ρ = −2/3, −1/3, 0, 1/3, 2/3, 1. Display each simulation in a separate
    scatterplot.

    Challenging Problems
    1. Let Xi , i = 1, 2, 3 be Γ(α, 1) random variables.
    (a) Find the density of S = X! + X2 + X3
    (b) Find the joint density of Yi = Xi /S, i = 1, 2.
    (c) Find the density of Y1 + Y2
    2. Let X ∼ Γ(α, β) and Y |X ∼ P ois(X)
    (a) Find E[Y |X] and Var(Y |X).
    Since Y |X ∼ P ois(X), E[Y |X] = Var(Y |X) = X
    (b) Find Var(Y ).
    V ar(Y ) = E(V ar(Y |X)) + V ar(E(Y |X))
    = EX + V ar(X) (7)
    α α
    = + 2
    β β

    2
    Figure 1: rho=-2/3

    Figure 2: rho=-1/3

    3
    Figure 3: rho=0

    Figure 4: rho=1/3

    4
    Figure 5: rho=2/3

    Figure 6: rho=1

    5
    Figure 7: codes

    (c) Find the density of Y . What family of random variables contains Y .


    Z ∞
    f (y) = f (y|x)f (x)dx
    0
    Z ∞ α−1
    x x e−x xy
    = α
    exp(− ) dx
    0 β Γ(α) β y!
    Z ∞
    1 β+1
    = xy+α−1 exp(− x)dx (8)
    y!β α Γ(α) 0 β
    Γ(α + y)β y
    =
    y!(β + 1)α+y Γ(α)
    Γ(α + y) β
    = exp(yln( ) − αln(1 + β))
    y!Γ(α) 1+β

    So this belongs to exponential family.

    You might also like

    pFad - Phonifier reborn

    Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

    Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


    Alternative Proxies:

    Alternative Proxy

    pFad Proxy

    pFad v3 Proxy

    pFad v4 Proxy