Professional Documents
Culture Documents
of Variance and Design of
Experiments
Results from Matrix Theory and Random Variables
:::
Lecture 2
Random Vectors and Linear Estimation
Shalabh
Department of Mathematics and Statistics
Indian Institute of Technology Kanpur
Slides can be downloaded from http://home.iitk.ac.in/~shalab/sp1
Random vectors:
Let Y1 , Y2 ,..., Yn be n random variables then Y (Y1 , Y2 ,..., Yn ) ' is
called a random vector.
• The mean vector Y is
E (Y ) (( E (Y1 ), E (Y2 ),..., E (Yn )) '.
3
Linear function of random variable:
If Y1 , Y2 ,..., Yn are n random variables, and k1 , k2 ,.., kn are scalars,
n
n
If Y (Y1 , Y2 ,..., Yn ) ', K (k1 , k2 ,..., kn ) ' then K ' Y kiYi ,
i 1
n
• the mean K ' Y is E ( K ' Y ) K ' E (Y ) ki E (Yi ) and
i 1
4
Multivariate normal distribution:
A random vector Y (Y1 , Y2 ,..., Yn ) ' has a multivariate normal
distribution with mean vector ( 1 , 2 ,..., n ) ' and dispersion
matrix if its probability density function is
1 1
f (Y | , ) 1/2
exp
2 (Y ) ' 1
(Y )
(2 ) n /2
5
Chi‐square distribution:
• If Y1 , Y2 ,..., Yk are independently distributed random variables
following the normal distribution with common mean 0 and
k
i 1
2
• The probability density function of ‐distribution with k
degrees of freedom is given as
1 x
k
1
f 2 ( x) x exp ; 0 x .
2
(k / 2)2 k /2
2
6
Chi‐square distribution:
• If Y1 , Y2 ,..., Yk are independently distributed following the
normal distribution with common mean 0 and common
1 k
Yi 2 2
, 2
variance then has distribution with k degrees
2 i 1
of freedom.
i 1
has non‐central 2
2 i
parameter 2
.
i 1
8
Chi‐square distribution:
• If U has a noncentral Chi‐square distribution with k degrees
of freedom and noncentrality parameter then
E U k and Var (U ) 2k 4 .
10
Chi‐square distribution:
Let the two quadratic forms‐
X ' A1 X is distributed as with n1 degrees of freedom
2
•
and noncentrality parameter ' A1 and
• X ' A2 X is distributed as 2
with n2 degrees of freedom
and noncentrality parameter ' A2 . .
Then X ' A1 X and X ' A2 X are independently distributed if
A1 A2 0.
11
t‐ distribution:
• If
X has a normal distribution with mean 0 and variance 1,
Y has a 2 distribution with n degrees of freedom, and
X and Y are independent random variables,
X
then the distribution of the statistic T is called the t‐
Y /n
distribution with n degrees of freedom.
13
F ‐ distribution:
• If X and Y are independent random variables with ‐
2
m /2
m n m
m2
mn
2 n m 2
fF ( f ) f 2
1 n f ; 0 f .
m n
2 2
14
F ‐ distribution:
• If X has a noncentral Chi‐square distribution with m degrees
of freedom and noncentrality parameter ; Y has a
2 distribution with n degrees of freedom, and X and Y are
15
Linear model:
Suppose there are n observations. In the linear model, we assume
that these observations are the values taken by n random
variables Y1 , Y2 ,..., Yn satisfying the following conditions:
1. E (Yi ) is a linear combination of p unknown parameters 1 , 2 ,..., p
E (Yi ) xi11 xi 2 2 ... xip p , i 1, 2,..., n
16
Linear model:
The linear model can be rewritten by introducing independent
normal random variables following N (0, 2 ) , as
Yi xi11 xi 2 2 .... xip p i , i 1, 2,..., n.
These equations can be written using the matrix notations as
Y X
• Y is a n x 1 vector of observation,
• X is a n x p matrix of n observations on each of
X 1 , X 2 ,..., X p
variables,
• is a p 1 vector of parameters and
• is a vector of random error components with
n 1
~ N (0, 2 I n ). 17
Linear model:
Here
• Y is called study or dependent variable,
• X 1 , X 2 ,..., X p are called explanatory or independent variables and
• 1 , 2 ,..., p are called as regression coefficients.
Alternatively since Y ~ N ( X , 2 I ) so the linear model can also be
expressed in the expectation form as a normal random variable Y
E (Y ) X
with
Var (Y ) 2 I .
2
Note that and are unknown but X is known.
18
Estimable functions:
A linear parametric function ' of the parameter is said to be
an estimable parametric function or estimable if there exists a
linear function of random variables 'Y of Y where Y (Y1 , Y2 ,..., Yn ) '
such that
E ( ' Y ) '
with (1 , 2 ,..., n ) ' and (1 , 2 ,..., n ) ' being vectors of
known scalars.
19
Best linear unbiased estimates (BLUE):
The unbiased minimum variance linear estimate 'Y of an
estimable function
'
is called the best linear unbiased
estimate (BLUE) of ' .
• Suppose 1' Y and '2Y are the BLUE of 1' and 2'
respectively.
(a11 a2 2 ) ' Y (a11 a2 2 ) ' .
Then is the BLUE of
• If ' is estimable, its best estimate is ' ˆ where ˆ
is any solution of the equations X ' X X ' Y .
20
Least squares estimation:
The least‐squares estimate of is Y X is the value
of which minimizes the error sum of squares ' .
Let S ' (Y X ) '(Y X )
Y 'Y 2 ' X 'Y ' X ' X .
Minimizing S with respect to involves
S
0
X ' X X 'Y
which is termed as normal equation.
This normal equation has a unique solution given by,
ˆ ( X ' X ) 1 X ' Y
assuming rank(X) = p.
21
Least squares estimation:
2S
Note that X ' X is a positive definite matrix.
'
• E (Y X ˆ ) 0.
23
Linear model with correlated observations:
In the linear model Y X with E ( ) 0, Var ( ) and
is normally distributed, we find
E (Y ) X , Var (Y ) .
Then
'Y ' X
~ N ,1 .
' '
( ' y ) 2
Further, 2 has a noncentral Chi‐square distribution with
'
( ' X ) 2
one degree of freedom and noncentrality parameter .
'
2
25
Degrees of freedom:
A linear function l’Y of the observations ( 0) is said to carry
one degrees of freedom.
A set of linear functions L’Y where L is r x n matrix, is said to have
M degrees of freedom if there exist M linearly independent
functions in the set and no more.
27
Sum of squares:
The sum of squares and the degrees of freedom arising out of
the mutually orthogonal sets of functions can be added
together to give the sum of squares and degrees of freedom for
the set of all the function together and vice versa.
28
Sum of squares:
Let X ( X 1 , X 2 ,..., X n ) has a multivariate normal distribution with
mean vector and positive definite covariance matrix .
Let the two quadratic forms.
• X ' A1 X is distribution 2 with n1 degrees of freedom and
noncentrality parameter ' A1 and
• X ' A2 X is distributed as 2 with n2 degrees of freedom and
noncentrality parameter ' A2 .
29
Fisher‐Cochran theorem:
If X ( X 1 , X 2 ,..., X n ) ' has a multivariate normal distribution with
mean vector and positive definite covariance matrix and
let X ' 1 X Q1 Q2 ... Qk
k
' ' Ai .
1
i 1
30
Derivatives of quadratic and linear forms:
Let X ( x1 , x2 ,..., xn ) ' and f(X) be any function of n
independent variables x1 , x2 ,..., xn , then
f ( X )
x
1
f ( X )
f ( X )
x2 .
X
f ( X )
x
n
K ' X
If K (k1 , k2 ,..., kn ) ' is a vector of constants, then K.
X
X ' AX
If A is a m n matrix, then 2( A A ') X .
X
31
Independence of linear and quadratic forms:
• Let Y be an n 1 vector having multivariate normal
distribution N ( , I ) and B be a m n matrix.
m1
Then the vector linear form BY is independent of the
quadratic form Y ' AY if BA = 0 where A is a symmetric
matrix of known elements.
• Let Y be a n 1 vector having multivariate normal
distribution N ( , ) with rank () n.
If BA 0, then the quadratic form Y ' AY is independent
of linear form BY where B is a m n matrix.
32