0% found this document useful (0 votes)
29 views

Lect 5 PDF

1) Functions of one random variable Y can be defined as transformations of another random variable X through some function g(X). 2) If X is a random variable, then Y = g(X) is also a random variable. 3) To find the distribution and density functions of Y, we must determine the sets on the x-axis such that X is mapped to values less than or equal to y through the function g, and take the probabilities of these sets.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
29 views

Lect 5 PDF

1) Functions of one random variable Y can be defined as transformations of another random variable X through some function g(X). 2) If X is a random variable, then Y = g(X) is also a random variable. 3) To find the distribution and density functions of Y, we must determine the sets on the x-axis such that X is mapped to values less than or equal to y through the function g, and take the probabilities of these sets.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 21

Functions of one Random Variable

1 aX + b
X X2

sin X Y = g(X ) |X |

X
log X
eX | X |U ( x)
Functions of a Random Variable

X: a r.v defined on the model (  , F , P ),


g(X): a function of the variable x, Y = g ( X ).

▪ Is Y necessarily a r.v?
▪ If so what is its PDF FY ( y ), pdf fY ( y ) ?
Functions of a Random Variable
▪ If X is a r.v, so is Y, and for every Borel set B,
P (Y  B ) = P ( X  g −1 ( B )).

▪ With Y ( )  B , In particular

FY ( y ) = P (Y ( )  y ) = P ( g ( X ( ))  y )
( ) (
= P X ( )  g −1 ( y ) = P X ( )  g −1 ( −, y ] .)
▪ To obtain the distribution function of Y, we must determine the Borel set on
the x-axis such that X ( )  g −1 ( y ) for every given y, and the probability of
that set.
Example: Y = aX + b

▪ For a  0 : ▪ For a  0,

FY ( y ) = P(Y ( )  y ) FY ( y ) = P (Y ( )  y )
= P(aX ( ) + b  y ) = P(aX ( ) + b  y )
 y −b  y −b
= P X ( )   = P X ( )  
 a   a 
 y −b  y −b
= FX  . = 1 − FX   ,
 a   a 

1  y −b 1  y−b
And f Y ( y ) = fX  . and hence fY ( y ) = − fX  .
a  a  a  a 

1  y−b
We conclude: fY ( y ) = fX  .
|a |  a 
Example: Y = X 2

FY ( y ) = P (Y ( )  y ) = P ( X 2 ( )  y ) .
▪ For y  0,

X 2
( )  y  =  ,

FY ( y ) = 0, y  0.

▪ For y  0, according to the figure, Y = X2


y

▪ the event {Y ( )  y} = { X ( )  y}
2

is equivalent to { x1  X ( )  x 2 }. x1 x2 X
Example: Y = X 2 - continued

FY ( y ) = P ( x1  X ( )  x2 ) = FX ( x2 ) − FX ( x1 )
= FX ( y ) − FX ( − y ), y  0.

 1
 ( )
f ( y ) + f X (− y ) ,
fY ( y ) =  2 y X
y  0,
 0, otherwise .

▪ If f X ( x ) represents an even function,

fY ( y ) =
1
y
fX ( y ) U ( y ).
Example: Y = X 2 - continued

1 − x2 / 2
▪ If X ~ N ( 0 ,1), so that f X ( x ) = e ,
2

▪ We obtain the p.d.f of Y = X 2 to be

1
fY ( y ) = e − y / 2U ( y ).
2y

▪ which represents a Chi-square r.v with n = 1, since  (1 / 2 ) =  .

▪ Thus, if X is a Gaussian r.v with  = 0, then Y = X2 represents a


Chi-square r.v with one degree of freedom (n = 1).
Example

 X − c, X  c, g( X )
▪ For: 
Y = g ( X ) =  0, − c  X  c, −c
X
 X + c,
c
 X  − c.

▪ Find the distribution and density function wrt Y.


Example

 X − c, X  c, g( X )
▪ For: 
Y = g ( X ) =  0, − c  X  c, −c
X
 X + c,
c
 X  − c.

▪ We have P (Y = 0 ) = P ( − c  X ( )  c ) = FX ( c ) − FX ( − c ).

▪ For y  0 , we have x  c , and Y ( ) = X ( ) − c so that

FY ( y ) = P (Y ( )  y ) = P ( X ( ) − c  y )
= P ( X ( )  y + c ) = FX ( y + c ), y  0.

▪ For y  0, we have x  − c , and Y ( ) = X ( ) + c so that

FY ( y ) = P (Y ( )  y ) = P ( X ( ) + c  y )
= P ( X ( )  y − c ) = FX ( y − c ), y  0.
Example – continued

▪ Thus,

 f X ( y + c ), y  0,

fY ( y ) = [ FX ( c ) − FX ( − c )] ( y ),
 f ( y − c ), y  0.
 X

FX ( x ) FY ( y )
g( X )
−c
c X x y
Continuous Functions of a Random Variable

▪ A continuous function g(x)


▪ g (x ) nonzero at all but a finite number of points
▪ has only a finite number of maxima and minima
▪ eventually becomes monotonic as | x |→  .
▪ Consider a specific y on the y-axis, and a positive increment y

g ( x)
y + y
y

x
x1 x1 + x1 x3 x3 + x3
x2 +  x2 x2
Continuous Functions of a Random Variable
▪ For f Y ( y ) Y = g ( X ) where g () is of continuous type,
y + y
P y  Y ( )  y +  y  =  f Y ( u ) du  f Y ( y )   y .
y

▪ y = g ( x ) has three solutions x1 , x2 , x3 when y  Y ( )  y +  y ,

▪ X could be in any one of three disjoint intervals: g ( x)

{x1  X ( )  x1 + x1}, y + y

{x2 + x2  X ( )  x2 }
y

x
or {x3  X ( )  x3 + x3 } . x1 x1 + x1 x3 x3 + x3
x2 +  x2 x2

So,
Py  Y ( )  y + y  = P{ x1  X ( )  x1 + x1}
+ P{ x2 + x2  X ( )  x2 } + P{ x3  X ( )  x3 + x3} .
Continuous Functions of a Random Variable

▪ For small y , xi , we get


f Y ( y )  y = f X ( x1 )  x1 + f X ( x 2 )( −  x 2 ) + f X ( x3 )  x3 .

▪ Here,  x1  0,  x2  0 and x3  0, so


|  xi | 1
fY ( y ) =  f X ( xi ) = f X ( xi )
i y i  y /  xi

▪ As y → 0,
1 1
fY ( y ) =  f X ( xi ) =  f X ( xi ).
i dy / dx x i g ( xi )
i

▪ If the solutions are all in terms of y, the right side is only a function of y.
Example Y = X 2
Revisited

▪ For all y  0, x1 = − y and x2 = + y Y = X2


y

▪ f Y ( y ) = 0 for y  0 . X
x1 x2
dy dy
▪ Moreover = 2 x so that =2 y
dx dx x = xi

1
▪ and using f Y ( y ) =  f X ( xi ) ,
i 
g ( xi )

 1
 (
f ( y ) + f X (− y ) ,
fY ( y ) =  2 y X
)
y  0,

 0, otherwise ,

which agrees with previous solution.


1
Example: Y =
X

▪ Find fY ( y ).

▪ Here for every y, x1 = 1 / y is the only solution, and

dy 1 dy 1
= − 2 so that = 2
= y 2
,
dx x dx x = x1 1/ y

1
▪ and substituting this into fY ( y ) =  f X ( xi ) , we obtain
i g ( xi )

1 1
fY ( y ) = 2 f X  .
y  y
Example

▪ Suppose f X ( x ) = 2 x /  2 , 0  x   , and Y = sin X . Determine fY ( y ).

▪ X has zero probability of falling outside the interval ( 0,  ).


▪ y = sin x has zero probability of falling outside ( 0,1).
▪ f Y ( y ) = 0 outside this interval.

▪ For any 0  y  1, the equation y = sin x has an infinite number of


solutions  , x1 , x2 , x3 ,  , where x1 = sin −1 y is the principal solution.

▪ using the symmetry we also get x2 =  − x1 etc. Further,


dy
= cos x = 1 − sin 2 x = 1 − y 2
dx
▪ so that
dy
= 1 − y2 .
dx x = xi
Example – continued
f X ( x)

x3
x
(a) 
y = sin x

x
x −1 x1 x2 x3

(b)

+
▪ for 0  y  1, 1
fY ( y ) =  f X ( xi ).
i =− 1− y 2
i 0

▪ In this case f X ( x−1 ) = f X ( x3 ) = f X ( x4 ) =  = 0


▪ (Except for f X ( x1 ) and f X ( x2 ) the rest are all zeros).
Example – continued

▪ Thus,

 2 x1 2 x2 
fY ( y ) =
1
( f X ( x1 ) + f X ( x2 ) ) =
1
 2 + 2 
1− y 2
1− y  
2  
 2
2( x1 +  − x1 )  , 0  y  1,
= =  1 − y 2
 2 1 − y2  0, otherwise.

fY ( y )

2
 y
1
Example: Y = tan X , X  (−  / 2,  / 2 ) .

▪ As x moves in (−  / 2,  / 2 ) , y moves in (−  , +  ) .
▪ The function Y = tan X is one-to-one for −  / 2  x   / 2 .
f X ( x)
−1
▪ For any y, x1 = tan y is the principal solution.
x
− /2  /2

dy d tan x
= = sec 2 x = 1 + tan 2 x = 1 + y 2 y = tan x
dx dx
y
− /2
x
1 1/  x1  / 2
fY ( y ) = f X ( x1 ) = , −   y  +
| dy / dx | x = x1 1+ y 2

(Cauchy density function with parameter equal to unity)

1
fY ( y ) =
1 + y2

y
Functions of a Discrete-type R.V

▪ Suppose X is a discrete-type r.v with


P ( X = xi ) = pi , x = x1 , x 2 ,  , xi , 
▪ and Y = g ( X ).

▪ Clearly Y is also of discrete-type, and when x = xi , y i = g ( xi ),


and for those yi s,

P (Y = y i ) = P ( X = xi ) = pi , y = y1 , y 2 ,  , yi , 
Example

▪ Suppose X ~ P (  ), so that

− k
P( X = k ) = e , k = 0,1,2, 
k!

▪ Define Y = X 2 + 1 . Find the p.m.f of Y.

▪ Solution: X takes the values 0,1,2 ,  , k , 


▪ Y only takes the values 1, 2, 5, , k 2 + 1,

P (Y = k 2 + 1) = P ( X = k )

▪ so that for j = k 2 + 1
 j −1
(
P (Y = j ) = P X = )
j − 1 = e−
( j − 1)!
, j = 1, 2, 5, , k 2 + 1, .

You might also like