Scribd
Upload
25 views6 pages

2.3 Conditional Distributions and Expectations

1. The document discusses conditional distributions and expectations. It provides definitions for conditional variance and explores how conditional variance relates to total variance. 2. An example is provided of a joint probability density function for random variables X and Y. Conditional distributions and expectations are derived and used to show the relationships between conditional variance, expected conditional variance, and total variance. 3. A second example is given involving a joint probability distribution table for random variables X and Y. Conditional distributions, expectations, and variances are calculated and used to further illustrate the relationships outlined in the document.

Uploaded by

d
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
25 views6 pages

2.3 Conditional Distributions and Expectations

1. The document discusses conditional distributions and expectations. It provides definitions for conditional variance and explores how conditional variance relates to total variance. 2. An example is provided of a joint probability density function for random variables X and Y. Conditional distributions and expectations are derived and used to show the relationships between conditional variance, expected conditional variance, and total variance. 3. A second example is given involving a joint probability distribution table for random variables X and Y. Conditional distributions, expectations, and variances are calculated and used to further illustrate the relationships outlined in the document.

Uploaded by

d
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/ 6

STAT 410

2.3

Fall 2016

Conditional Distributions and Expectations.

(continued)

Var ( X | Y ) = E [ ( X E ( X | Y ) ) 2 | Y ] = E ( X 2 | Y ) [ E ( X | Y ) ] 2

Def

Theorem

E( E( X | Y ) ) = E( X )
Var ( E ( X | Y ) ) Var ( X )
Var ( X ) = Var ( E ( X | Y ) ) + E [ Var ( X | Y ) ]

Furthermore,

Var ( X )

Var ( E ( X | Y ) )

Portion of the
distribution of X
that is related to Y

E [ Var ( X | Y ) ]
Portion of the
distribution of X
that is independent
of Y

If X is a function of Y, then
E( X | Y ) = X
Var ( X | Y ) = 0

Var ( E ( X | Y ) ) = Var ( X ),

and
and

E [ Var ( X | Y ) ] = 0.

If X and Y are independent, then


E( X | Y ) = E( X )
Var ( X | Y ) = Var ( X )

Var ( E ( X | Y ) ) = 0 since E ( X ) is a constant,

and
and

E [ Var ( X | Y ) ] = Var ( X ).

2.

Let the joint probability density function for ( X , Y ) be


2

f (x, y ) = 60 x y 0 x 1, 0 y 1, x + y 1

Recall:

otherwise

f X ( x ) = 30 x 2 ( 1 - x ) 2 ,
E( X ) =

1
,
2

9
,
252

Var ( X ) =

f Y ( y ) = 20 y ( 1 - y ) 3 ,
E( Y ) =

0 < x < 1,

1
,
3

E(X2) =

2
72
= ;
252 7

E(Y2) =

1
36
= ;
252 7

0 < y < 1,

8
,
252

Var ( Y ) =

f Y|X ( y | x ) =

2y
f (x,y )
=
,
f X (x )
(1 - x )2

0 < y < 1 x,

0 < x < 1.

f X|Y ( x | y ) =

f (x,y )
3x2
=
,
3
f Y ( y)
(1 - y )

0 < x < 1 y,

0 < y < 1.

E( Y | X = x ) =

1- x

(1- x )

dy =
2

2
(1 - x ),
3

2
( 1 - X ).
3

E( Y | X ) =

E[ ( E( Y | X ) )2 ] =

Var ( E ( Y | X ) ) =

2y

E( E( Y | X ) ) =

( )) =

4
1 - 2 E( X ) + E X 2
9

8 1
-
63 3

0 < x < 1.

1
= E ( Y ).
3

4
1 2
8
.
1- 2 + =
9
2 7
63

1
.
63
OR

4
4 9
4
1
2

Var ( E ( Y | X ) ) = Var ( 1 - X ) =
=
=
.
Var ( X ) =
9
9 252
252
63
3

Var ( E ( Y | X ) ) =

E( Y | X = x ) =

1
4
8
=
<
= Var ( Y ).
63
252
252

1- x

2
y

2y

(1 - x )

dy =

1
(1 - x )2 ,
2

0 < x < 1.

Var ( Y | X = x ) = E ( Y 2 | X = x ) [ E ( Y | X = x ) ] 2
=
Var ( Y | X ) =

1
(1 - x )2 4 (1 - x )2 = 1 (1 - x )2 ,
2
9
18

0 < x < 1.

1
(1 - X )2 .
18

E [ Var ( Y | X ) ] =

( )) =

1
1 - 2 E( X ) + E X 2
18

Var ( E ( Y | X ) ) + E [ Var ( Y | X ) ] =

1
1 2
1
.
1- 2 + =
18
2 7
63

1
1
8
=
= Var ( Y ).
+
63 63
252

1- y

E( X | Y = y ) =

E( X | Y ) =

3x2

(1 - y )

dx =
3

3
( 1 - y ),
4

0 < y < 1.

3
( 1 - Y ).
4

E( E( X | Y ) ) =

3
( 1 - E ( Y ) ) = 3 1 - 1 = 1 = E ( X ).
2
4
3
4

9
9 8
1
3

Var ( E ( X | Y ) ) = Var ( 1 - Y ) =
=
.
Var ( Y ) =

16
16 252
56
4

Var ( E ( X | Y ) ) =

1
4.5
9
=
<
= Var ( X ).
56
252
252

1- y

E( X | Y = y ) =

x2

3x2

(1- y )

dx =

3
(1 - y )2 ,
5

0 < y < 1.

Var ( X | Y = y ) = E ( X 2 | Y = y ) [ E ( X | Y = y ) ] 2
=
Var ( X | Y ) =

3
(1 - y )2 9 (1 - y )2 = 3 (1 - y )2 ,
5
16
80

0 < y < 1.

3
(1 - Y )2 .
80

E [ Var ( X | Y ) ] =

( )) =

3
1 - 2 E( Y ) + E Y 2
80

Var ( E ( X | Y ) ) + E [ Var ( X | Y ) ] =

3
1 1
1
.
1- 2 + =
80
3 7
56

1
1
9
=
= Var ( X ).
+
56 56
252

1.

Consider the following joint probability distribution p ( x, y ) of two random


variables X and Y:

x
1
2

pX( x )

0.15

0.10

0.25

0.25

0.30

0.20

0.75

pY( y )

0.40

0.40

0.20

p X|Y ( x | 0 )

/0.40 = 0.375
0.25
/0.40 = 0.625

0.15

p X|Y ( x | 1 )

/0.40 = 0.25
0.30
/0.40 = 0.75

0.10

p X|Y ( x | 2 )

/0.20 = 0.00
0.20
/0.20 = 1.00
0.00

E ( X | Y = 0 ) = 1.625

E ( X | Y = 1 ) = 1.75

E( X | Y = 2 ) = 2

Var ( X | Y = 0 ) = 0.234375

Var ( X | Y = 1 ) = 0.1875

Var ( X | Y = 2 ) = 0

Def

Var ( X | Y ) ) = E [ ( X E ( X | Y ) ) 2 | Y ] = E ( X 2 | Y ) [ E ( X | Y ) ] 2

E( X | Y = y )

pY( y )

Var ( X | Y = y )

pY( y )

1.625

0.40

0.234375

0.40

1.75

0.40

0.1875

0.40

0.20

0.20

E ( E ( X | Y ) ) = 1.75 = E ( X )
Var ( E ( X | Y ) ) = 0.01875

<

E ( Var ( X | Y ) ) = 0.16875
0.1875 = Var ( X ).

Var ( E ( X | Y ) ) + E ( Var ( X | Y ) ) = 0.01875 + 0.16875 = 0.1875 = Var ( X ).

p Y|X ( y | 1 )

0.15

/0.25 = 0.60

0.25

0.10

/0.25 = 0.40
0.00
/0.25 = 0.00

0.30

/0.75 = 5/15

/0.75 = 6/15
0.20
/0.75 = 4/15

E ( Y | X = 1 ) = 0.4 = 6 15

E ( Y | X = 2 ) = 14 15

Var ( Y | X = 1 ) = 0.24 = 54 225

p Y|X ( y | 2 )

Var ( Y | X = 2 ) = 134 225

E( Y | X = x )

pX( x )

Var ( Y | X = x )

pX( x )

/15

0.25

0.25

/15

0.75

/225
134
/225

14

54

E ( E ( Y | X ) ) = 12 15 = 0.80 = E ( Y )

Var ( E ( Y | X ) ) = 4 75 0.053333

<

0.75

E ( Var ( Y | X ) ) = 38 75

0.56 = Var ( Y )

Var ( Y ) = 0.56 = 4 75 + 38 75 = Var ( E ( Y | X ) ) + E ( Var ( Y | X ) )

You might also like