Scribd Logo
Upload

Welcome to Scribd!

  • 1 views

    Ie265 8 241116 182540

    Uploaded by

    Halil Emre

    Copyright:

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

    Ie265 8 241116 182540

    Uploaded by

    Halil Emre
    1 views22 pages

    Original Title

    IE265_8_241116_182540

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    1 views22 pages

    Ie265 8 241116 182540

    Uploaded by

    Halil Emre

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 22

    IE 265

    • Functions of Random Variables


    2

    Discrete function of a discrete RV


    • Suppose discrete random variable (RV) X has pmf p(x).
    Given a mapping from X to discrete RV Y, Y = H(X), how
    can we find the pmf of Y?

    x1  p ( x1 )  p ( x 2 ), y  y1
    x2 y1 
     p ( x 3 )  p ( x 5 ), y  y2
    x3 p( y)  
    y2 y  y3
     p ( x 4 ),
    x4 y3  0 , otherwise
    x5
    3

    Discrete function of a discrete RV


    • Ex: Given the pmf of X below, find the pmf of
    Y  H ( X ) | X  2 |

    1 / 8, x0

    3 / 8, x 1
    
    p ( x )   3 / 8, x2
    1 / 8, x3

     0, otherwise
    4

    Discrete function of a continuous RV


    • Suppose continuous RV X has pdf f(x). Given a mapping
    from X to discrete RV Y, Y = H(X), how can we find the
    pmf of Y? x
    2
      f ( x ) dx , y  y1
     x1
    x4  x3
     f ( x ) dx ,
    p ( y )   x2
    y3  y  y2
    x3
    x2 y2  x4

      f ( x ) dx , y  y3
    y1
    x1
     x3
     0 , otherwise
    5

    Discrete function of a continuous RV


     e  x , x0
    • Ex: Given the pdf of X, f ( x )  
    0, x0

    0, X  1/ 
    find the pmf of Y  H ( X )  
    1, X  1/ 
    6

    Continuous function of a continuous RV


    Suppose continuous RV X has pdf f(x). Given a continuous
    mapping, Y = H(X), how can we find the pdf of Y (which is
    naturally continuous)?

    Procedure:
    1. Find the cdf of Y, F(y), using the mapping and f(x).
    2. Differentiate F(y) to find f(y).
    3. Find the range of Y.
    7

    Continuous function of a continuous RV


     x / 8, 0 x4
    • Ex: Given the pdf of X, f ( x )  
    0, otherwise

    find the pdf of Y  H ( X )  2 X  8


    8

    Continuous function of a continuous RV


    • Theorem: If X is a continuous RV with pdf f(x) > 0 for
    a < x < b, and if y = H(x) is a continuous strictly increasing
    or strictly decreasing function of x, then the RV Y = H(X)
    has pdf
    dx
    f ( y)  f ( x)
    dy

    with x=H-1(y) expressed in terms of y.

    The theorem, which is valid for the underlined special


    case of H(X), eliminates the need to find F(y) in step 1 of
    the procedure (particularly useful when integration is
    difficult).
    9

    Continuous function of a continuous RV


     x / 8, 0 x4
    • Ex: f ( x )  
    0, otherwise

    Y  H ( X )  2 X  8 is a strictly increasing function of X


    10

    Expected value of a function of a RV


    Theorem: If h(x) is a real valued function of x
    • Discrete case: If X has pmf p(x), then
    E [ h ( X )]   h ( x ) p ( x )
    x

    • Continuous case: If X has pdf f(x), then



    E [ h ( X )]   h ( x ) f ( x ) dx
    
    11

    Expected value of a function of a RV


    • Ex: The # of defectives in a batch of parts has pmf

     0 . 3, x 1
     0 . 5, x2

    p( x)  
     0 .2 , x3
     0 , otherwise

    If the cost of having x defectives is y = h(x) = 2x2 + 1, what


    is the expected cost of defectives?
    12

    Expected value of a function of a RV


    1 x / 2
     e , x0
    • Ex: Given the pdf of X, f ( x )   2
     0 , x0

    find the expected value of Y  h ( X )  5 X  3


    13

    Properties of mean and variance


    • For any random variable X and H(X) = aX + b, where a
    and b are any constants:
    1. E ( aX  b )  aE ( X )  b
    Proof: E ( aX  b )   ( ax  b ) p ( x )
    x
    (similar for
      axp ( x )   bp ( x ) continuous
    x x

     a  xp ( x )  b  p ( x )
    case)
    x x

     aE ( X )  b

    Hence E(b) = b, E(aX) = aE(X).


    14

    Properties of mean and variance


    2. Var ( X )  E ( X 2 )  [ E ( X )] 2
    Proof: Var ( X )   [ x  E ( X )] 2 p ( x )
    x

       
    2 2
    { x 2 xE ( X ) [ E ( X )] } p( x)
    x

      x 2 p ( x )  2 E ( X )  xp ( x )  [ E ( X )] 2  p ( x )
    x x x

    E(X 2) E(X ) 1
     E ( X 2 )  [ E ( X )] 2
    15

    Properties of mean and variance


    3. Var ( aX  b )  a Var ( X )
    2

    Proof: Var ( aX  b )  E {[( aX  b )  E ( aX  b )] 2 }


     E {( aX  b ) 2  2 ( aX  b ) E ( aX  b )  [ E ( aX  b )] 2 }
    ( 2 aX  2 b )[ aE ( X )  b ] [ aE ( X )  b ] 2
     E {a 2 X 2  2 abX  b 2
     2 a 2 XE ( X )  2 abX  2 abE ( X )  2 b 2
     a 2 [ E ( X )] 2  2 abE ( X )  b 2 }
     E {a 2 [ X 2  2 XE ( X )  ( E ( X )) 2 ]}
     a 2 E {[ X  E ( X )] 2 }
     a 2Var ( X )
    16

    Properties of mean and variance


    If X and Y are two independent random variables:
    4. E ( X  Y )  E ( X )  E (Y )
    5. Var ( X  Y )  Var ( X )  Var (Y )
    17

    Properties of mean and variance


    4 x 3 , 0  x 1
    • Ex: f ( x )  
    0, otherwise

    a) Find the variance of X using property 2.


    b) Find the expected value and variance of Y = 2X + 3.
    18

    Moment generating function


    • The rth moment of random variable X taken about the
    origin is E(Xr) (expected value of the rth power of X).

    • Moments are useful in calculating distribution


    characteristics such as mean and variance, particularly
    when closed form of a summation or integration cannot be
    obtained, e.g.
      E ( X ) first moment
     2  Var ( X )  E ( X 2 )  [ E ( X )] 2 second moment is used
    19

    Moment generating function


    • Moment generating function of X, E(etX), packages all
    the moments of a random variable in one simple
    expression
    M (t )  E ( e tX )   p ( x ),
    tX
    e if X is discrete
    x

      f ( x ) dx ,
    tX
    e if X is continuous
    

    • M(t) is used to find moments of X with the help of Taylor


    series expansion

    (tX ) i t2X 2 t3X 3 tr X r
    e tX    1  tX    
    i 0 i! 2! 3! r!
    20

    Moment generating function


    t2 t3 tr
    M (t )  E ( e tX )  1  tE ( X )  E(X 2)  E(X 3)   E(X r ) 
    2! 3! r!

    dM 2t 3t 2 rt r 1
     E(X )  E(X ) 2
    E(X )  
    3
    E(X r )   E(X )
    dt t 0 2! 3! r!

    d 2M 6t r ( r  1)t r  2
     E(X ) 
    2
    E(X )  
    3
    E(X r )   E(X 2)
    dt 2 t 0
    3! r!

    d rM
     E(X r )
    dt r t 0
    21

    Moment generating function


     n  x n x

       p (1  p ) , x  0,1, 2,  , n
    • Ex: p ( x )   x 

    0, otherwise

    Find Var(X) using moment generating function.


    22

    Moment generating function


    • Ex: An experiment has two outcomes such that
    P (success )  p and P ( failure )  1  p

    1, if outcome is success P ( X  1)  p


    X 
    0, if outcome is failure P ( X  0)  1  p

    Find Var(X) using moment generating function.

    You might also like