Matrix Algebra
properties of invertible matrix
(
A
B
)
−
1
=
B
−
1
A
−
1
(AB)^{-1}=B^{-1}A^{-1}
(AB)−1=B−1A−1
(
A
T
)
−
1
=
(
A
−
1
)
T
(A^T)^{-1}=(A^{-1})^{T}
(AT)−1=(A−1)T
[
A
I
]
→
[
I
A
−
1
]
[A\quad I] \rightarrow[I\quad A^{-1}]
[AI]→[IA−1]
condition number of matrix:
c
o
n
d
(
A
)
=
∣
∣
A
∣
∣
∗
∣
∣
A
−
1
∣
∣
cond(A)=||A||*||A^{-1}||
cond(A)=∣∣A∣∣∗∣∣A−1∣∣
big condition number means the matrix is in ill-condition and it maybe not invertible
Here are some statements either all true or all false:
a.A is an invertible matrix
b.A is row equavalent to
n
×
n
n\times n
n×n identity matrix
c.A has n pivot positions
d.The equation
A
x
=
0
Ax=0
Ax=0 has only the trivial solution
e.The columns
A
A
A from a linear identity set.
f. The linear transform
x
→
A
x
x\rightarrow Ax
x→Ax is one to one
g.The equation
A
x
=
b
Ax=b
Ax=b has at least one solution for each
b
b
b in
R
n
R^n
Rn
h.The columns of
A
A
A span
R
n
R^n
Rn
i.The linear transformation
x
→
A
x
x\rightarrow Ax
x→Ax maps
R
n
R^n
Rn onto
R
n
R^n
Rn
j.There is an
n
×
n
n\times n
n×nmatrix
C
C
C such that
C
A
=
I
CA=I
CA=I or
A
C
=
I
AC=I
AC=I
i.
A
T
A^T
AT is an invertible matrix
Partitioned Matrices
A matrix of the form A = [ A 11 A 12 0 A 22 ] A=\begin{bmatrix}A_{11}&A_{12} \\0&A_{22} \end{bmatrix} A=[A110A12A22] is said to be block upper triangular. Assume that A 11 A_{11} A11 is p × p p\times p p×p, A 22 A_{22} A22 is q × q q\times q q×q and A A A is invertible. Then A − 1 = [ A 11 − 1 − A 11 − 1 A 12 A 22 − 1 0 A 22 − 1 ] A^{-1}=\begin{bmatrix}A_{11}^{-1}& -A_{11}^{-1}A_{12}A_{22}^{-1}\\0&A^{-1}_{22}\end{bmatrix} A−1=[A11−10−A11−1A12A22−1A22−1]
Matrix Factorizations
LU Factorization
A
=
L
U
=
[
1
0
0
0
∗
1
0
0
∗
∗
1
0
∗
∗
∗
1
]
[
×
∗
∗
∗
∗
0
×
∗
∗
∗
0
0
0
×
∗
0
0
0
0
×
]
A=LU=\begin{bmatrix}1&0&0&0\\*&1&0&0\\*&*&1&0\\*&*&*&1\end{bmatrix} \begin{bmatrix}\times&*&*&*&*\\0&\times&*&*&*\\0&0&0&\times&*\\0&0&0&0&\times\end{bmatrix}
A=LU=⎣⎢⎢⎡1∗∗∗01∗∗001∗0001⎦⎥⎥⎤⎣⎢⎢⎡×000∗×00∗∗00∗∗×0∗∗∗×⎦⎥⎥⎤
Then the solution of
A
x
=
b
Ax=b
Ax=b could be get through
L
y
=
b
Ly=b
Ly=b and
U
x
=
y
Ux=y
Ux=y
Subspaces of R n R^n Rn
The subspaces of
R
n
R^n
Rn is any set
H
H
H in
R
n
R^n
Rn that has three propoerties:
a.The zero vector in
H
H
H.
b.For each
u
u
u and
v
v
v in
H
H
H, the sum
u
+
v
u+v
u+v is in
H
H
H
c.For each
u
u
u in
H
H
H and each scalar
c
c
c, the vector
c
u
cu
cu is in H
The Null space of a matrix A is the set Nul A of all solutions of the homogeneous equation
A
x
=
0
Ax=0
Ax=0
The Null space of a
m
×
n
m\times n
m×n matrix A is subspace of
R
n
R^n
Rn
Basis: A basis for a subspace H of
R
n
R^n
Rn is a linearly independent set in H that span H
The pivot columns of a matrix
A
A
A form a basis for the column space of
A
A
A.
Demension and Rank
Demension of a nonzero subspace
H
H
H, denoted by dim
H
H
H, is the number of vectors in any basis for
H
H
H. The demension of the zero subspace {0} is definded to be zero.
The rank of a matrix
A
A
A, denoted by rank
A
A
A, is the demension of the column space of
A
A
A.
If a matrix
A
A
A has n columns, then
r
a
n
k
A
+
d
i
m
N
u
l
A
=
n
rank A + dim Nul A = n
rankA+dimNulA=n
Rank and the Invertible matrix theorem
Let
A
A
A be
n
×
n
n\times n
n×n matrix. Then the following statements are each equivalent to the statement that
A
A
A is an invertible matrix:
m. the columns of
A
A
A form a basis of
R
n
R^n
Rn
n. Col A=
R
n
R^n
Rn
o. dim Col A=n
p. rank A=n
q. Nul A={0}
r. dim Nul A=0