Scribd Logo
Upload

Welcome to Scribd!

  • 8 views

    Supplement05-T Distribution

    Uploaded by

    erichhhhh
    1) The document shows that the joint distribution of independent normal and chi-squared random variables can be transformed into a joint t-distribution and gamma distribution. 2) Integrating the joint density over one variable gives the density of the remaining variable, which is shown to be a student's t-distribution. 3) Therefore, the document demonstrates that the student's t-distribution can be derived from the transformation of normal and chi-squared random variables.

    Copyright:

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

    Supplement05-T Distribution

    Uploaded by

    erichhhhh
    8 views1 page
    1) The document shows that the joint distribution of independent normal and chi-squared random variables can be transformed into a joint t-distribution and gamma distribution. 2) Integrating the joint density over one variable gives the density of the remaining variable, which is shown to be a student's t-distribution. 3) Therefore, the document demonstrates that the student's t-distribution can be derived from the transformation of normal and chi-squared random variables.

    Original Title

    supplement05-t_distribution

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    1) The document shows that the joint distribution of independent normal and chi-squared random variables can be transformed into a joint t-distribution and gamma distribution. 2) Integrating the joint density over one variable gives the density of the remaining variable, which is shown to be a student's t-distribution. 3) Therefore, the document demonstrates that the student's t-distribution can be derived from the transformation of normal and chi-squared random variables.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    8 views1 page

    Supplement05-T Distribution

    Uploaded by

    erichhhhh
    1) The document shows that the joint distribution of independent normal and chi-squared random variables can be transformed into a joint t-distribution and gamma distribution. 2) Integrating the joint density over one variable gives the density of the remaining variable, which is shown to be a student's t-distribution. 3) Therefore, the document demonstrates that the student's t-distribution can be derived from the transformation of normal and chi-squared 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 1

    Note: T Distribution

    March 19, 2020

    We first show that the joint density of independent random variables X and Y that
    follow the normal distribution N (0, 1) and chi-square distribution χ2 (m) respectively, is
    m
    1 1 (1) 2 m 1
    fZ,Y (z, y) = √ exp{− z 2 } 2 m y 2 −1 e− 2 y for y > 0. (1)
    2π 2 Γ( 2 )

    Then map (z, y) to (u, v) by u = y, v = √z y . The inverse transformations are y = u,


    pu m
    z=v m , and the Jacobian of the inverse is

    ∂z ∂y
    r
    ∂u ∂u
    u
    J= ∂z ∂y =− .
    ∂v ∂v
    m

    Now the joint density of U and V is

    fU,V (u, v) = |J|fZ,Y (z(u, v), y(u, v))


    m
    u 2 ( 12 ) 2 m −1 − 1 u
    r r
    u 1 1
    = √ exp{− (v ) } m u2 e 2
    m 2π 2 m Γ( 2 ) (2)
    m
    m+1
    −1 1 v2 1 ( 12 ) 2
    =u 2 e− 2 ( m +1) √ m .
    2πm Γ( 2 )
    m+1 1 v2
    And we can observe that u 2 −1 e− 2 ( m +1) follows the density function of the gamma
    1 v2
    distribution Ga( m+1
    2 , 2 ( m + 1)). We integrate over u to obtain :

    2 m+1 m
    ∞ ( 12 ( vm + 1)) Γ( m+1
    2 ) 1 ( 12 ) 2
    Z 2
    2 m+1
    −1 − 12 ( vm +1)
    fV (v) = u 2 e du ∗ √ m
    Γ( m+1
    m+1 2 m+1
    0 2 ) ( 12 ) 2 ( vm + 1)) 2 2πm Γ( 2 )
    Γ( m+1
    2 ) v2 m+1
    =1∗ √ m ( + 1)− 2
    πmΓ( 2 ) m
    Γ( m+1
    2 ) v2 m+1
    =√ m ( + 1)− 2 ∼ t(m)
    πmΓ( 2 ) m
    (3)
    We can see that fV (v) is the probability density function of t(m).

    You might also like