Determinants and Linear Dependence
1
2
3
Cramer’s Rule
●
This is a formula in terms of determinants for the solution of n
linear equations in n unknowns when there is exactly one solution
Cramer’s rule to solve two equations in two unknowns
Multiply the first equation by b2, the second by b1, and then
subtract the results and solve for x and y
4
Solutions for two unknowns x and y
determinant of the coefficients
5
Cramer’s rule for solving a system of three equations and three
unknowns
Given
then
6
Example: Solve the set of equations
7
Example: Solve the set of equations
Solution:
8
Revisiting the circuit problem using matrix
( R1 R2
R 1 +R 3 R3 )( ) (
I1
I2
=
V 1 −V 2
V 1 +V 3 )
9
10
Rank of a Matrix (Alternative Method)
●
A submatrix means a matrix remaining if we remove some rows and/or remove
some columns from the original matrix.
●
To find the rank of a matrix, we look at all the square submatrices and find their
determinants. The order of the largest nonzero determinant is the rank of the
matrix.
11
Example. Find the rank of the matrix
You need to look at the four 3 by 3 determinants containing
columns 1,2,3 or 1,2,4 or 1,3,4 or 2,3,4.
12
Solution
Note that the first two columns are negatives of each other, so by
Fact 2c the determinants containing columns 1,2,3 or 1,2,4 of are
both zero
The determinants containing column 1,3,4 or 2,3,4 differ only
in the sign of their first column, so we just have to look at one of
them
13
T-2F
by Fact 2c that the determinant is zero
So the rank of the matrix is less than 3. To show that it is 2,
we just have to find one 2 by 2 submatrix with nonzero
determinant.
You can show that the determinant of
some 2x2 submatrices is non-zero.
Thus the rank of the matrix is 2.
|2 3
−1 0 |
=(2⋅0)−(−1⋅3)=3
14
Matrix Operation
●
Multiplication of a Matrix by a Number
●
Addition of Matrices
●
Multiplication of Matrices
15
Multiplication of a Matrix by a Number
●
Operation is commutative: kA=Ak 16
For nxn matrix
det (kA )=k n det ( A )
17
Addition, Subtraction of Matrices
18
Multiplication of matrices (inner product)
●
Requires that A has the same number of rows as the number of column
of B
●
The product has the same number of rows of A and the same number of
columns of B 19
Multiplication of matrices
●
In general, the operation is non-commutative AB≠BA
●
Defn: Commutation Relation [ A , B ]=AB−BA
●
Matrix Multiplication is Associative ( AB)C= A ( BC)
20
Example: Pauli Matrices
21
Seatwork
If
find AB and BA.
22
Seatwork
If find AB and BA.
Soln
23
Operations with Determinants
A=
( a11 a12
a21 a22 ) B=
( b11 b12
b 21 b 22 ) C= AB=
( a 11 b11 +a 12 b21 a11 b 12 +a12 b 22
a 21 b11 +a 22 b21 a21 b 12 +a22 b 22 )
det ( A)=a 11 a22−a21 a12 det ( B)=b11 b 22−b21 b 12
det ( A) det (B)=a11 a 22 b11 b 22−a11 a22 b 21 b12 −a21 a12 b11 b 22 +a 21 a12 b21 b 12
det ( AB)=(a11 b11 + a12 b 21)(a21 b 12 +a 22 b22 )−(a 21 b11 + a22 b 21)(a11 b 12 +a12 b 22)
= a11 a22 b 11 b22 −a11 a 22 b21 b 12−a21 a12 b 11 b22 + a21 a12 b 21 b12
24
Unit Matrix (Identity Matrix): a square matrix with elements of
value unity on its principal diagonal (the elements (i,j) with i=j ), and
zeros everywhere else
( )
1 0 0 0 ... 0
0 1 0 0 ... 0
I = 0 0 1 0 ... 0
0 0 0 1 ... 0
0 0 0 0 ... 1
IA= AI = A
If A is mxn, the unit matrix in 1A must be mxm, while that in A1 must be nxn.
25
Diagonal Matrices: has nonzero elements dij only for i = j
( )
d 11 0 0 0... 0
0 d 22 0 0... 0
D= 0 0 d 33 0... 0
0 0 0 d 44 ... 0
0 0 0 0 ... d nn
The rules of matrix multiplication cause all diagonal matrices (of the same
size) to commute with each other. However, unless proportional to a unit
matrix, diagonal matrices will not commute with nondiagonal matrices
containing arbitrary elements.
26
Applications of Matrix Multiplication
27
Matrix Inverse:
The inverse of a matrix M (if it has one) as the matrix M−1 such that
MM−1 and M−1M are both equal to a unit matrix I.
●
only square matrices can have inverses (otherwise we could
not multiply both MM−1 and M−1M)
●
M−1 must be unique
●
det(M)≠ 0 is a requirement for M to have an inverse
●
If a matrix has an inverse we say that it is invertible; if it doesn’t
have an inverse, it is called singular.
28
Example: Find the inverse of M
29
Example: Find the inverse of det ( M )=3
30
Thus,
Going back with the linear eqns,
31
GAUSS-JORDAN MATRIX INVERSION (Alternative method)
The Gauss-Jordan method is based on the fact that there exist matrices ML
such that the product MLA will leave an arbitrary matrix A unchanged, except
with
●
(a) one row multiplied by a constant, or
●
(b) one row replaced by the original row minus a multiple of another row, or
●
(c) the interchange of two rows.
If A is nonsingular, the application of a succession of ML
●
Apply successive transformations to A until these transformations
have reduced A to 1, keeping track of the product of these
transformations
●
To keep track is to successively apply the transformations to a unit
32
matrix
Using GAUSS-JORDAN MATRIX INVERSION, Find the inverse of A
33
Using GAUSS-JORDAN MATRIX INVERSION, Find the inverse of A
Strategy: Write, side by side, the matrix A and a unit matrix of the same size,
and to perform the same operations on each until A has been converted to a unit
matrix, which means that the unit matrix will have been changed to A-1
34
35
36
37
38
Derivatives of Determinants
Differentiation with respect to the x dependence of its element aij ,
By chain rule,
39
Rotation Matrices
40
Rotation Matrices (beware I interchanged the angle!) (X’,Y’)
R’
41
Functions of Matrices
The constant term c or cA0 in a polynomial is defined to mean
c times the unit matrix I
42
43
44
Transpose
~ T
Transpose of A: A , A
A matrix that is unchanged by transposition (i.e., AT =A) is called symmetric
45
A matrix is symmetric if it satisfies the condition A = AT, i.e.
aij = aji, and anti-symmetric (or skew symmetric) if A = −AT, i.e.
aij = −aji, where AT is the transpose of A.
Any matrix A may be expressed as the sum of a symmetric and
an anti-symmetric matrix
Symmetric Anti-Symmetric
46
Adjoint
Complex conjugate i* =-i
The adjoint of a matrix A, denoted A†
(dagger), is obtained by both complex
conjugating and transposing it
47
Trace The trace, a quantity defined for square matrices, is the sum of
the elements on the principal diagonal. Thus, for an nxn matrix A,
(holds even if AB≠BA.
What is the trace of any commutating matrices, [A,B] = AB-BA =0 ?
48
Operations on Matrix Products
49
50
Matrix Representation of Vectors
() ()
a1 b1
Suppose a= a2 b= b 2
a3 b3
51
Orthogonal Matrices
the matrix product of two orthogonal matrices is also orthogonal
52
Unitary Matrices: matrices for which the adjoint is also the inverse
(if all the elements of a unitary matrix are real, the matrix is also orthogonal)
●
det(U) is a possibly complex number of magnitude unity
●
if U and V are both unitary, then UV and VU will be unitary
53
Hermitian Matrices
A (square) matrix is identified as Hermitian, or, synonymously, self-adjoint,
if it is equal to its adjoint
●
elements exhibits a reflection symmetry about the principal diagonal
●
diagonal elements must be real
●
if two matrices A and B are Hermitian, it is not necessarily true that AB or BA
is Hermitian
●
AB+BA, if nonzero, will be Hermitian, and AB-BA, if nonzero,
will be anti-Hermitian
54
55
If A† = −A, the matrix A is said to be anti-Hermitian (or skew
Hermitian)
Any complex matrix can be expressed as the sum of a Hermitian
matrix and an anti-Hermitian matrix
Hermitian anti-Hermitian
56
Example:
57
Solution:
58
59
60
61
62
63