Advanced Higher Notes (Unit Matrice

3) s

Matrices
Prerequisites: Adding, subtracting, multiplying and dividing numbers;
elementary row operations.

Maths Applications: Solving systems of equations; describing

geometric transformations; deriving addition
formulae.

Real-World Applications: Balancing chemical equations; flight

stopover information; currents in electrical
circuits; formulation of fundamental physical
laws.

Basic Definitions

Definition:

A matrix is a rectangular array of numbers (aka entries or elements)

in parentheses, each entry being in a particular row and column.

Definition:

The order of a matrix is given as m  n (read m by n), where m

is the number of rows and n the number of columns and is written
as,

 a11 a1 a1 … a1n 
a
2
a
3
… 
def
 a21 2n 
A  ( aij )   a31 … a3n 
22 23

 a32 a33 
mn  ⋯ % 
⋯ ⋯
a am 
 m am am 
1 n

The element in row i and column j of a matrix is written as aij and

called the (i, j) th
entry of A.

In this course, we will deal almost exclusively with matrices that

have orders 2  2 and 3  3.

M Patel (April 1 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
Definition:

The main diagonal (aka leading diagonal) of any matrix is the set
of entries aij where i = j.

Special cases arise when either m = 1 or n = 1.

Definition:

A row matrix is a 1  n matrix and is written as,

a 11
a12 … a1(n 1) a1n 
A column matrix is a m  1 matrix and is
written as,

 a1 
 1 
 a 
 ⋯2 
 
 a (m 1)1 
 am 
 
1

The case when m = n is a very important one.

Definition:

A square matrix (of order m  m) is a matrix with the same

number of rows as columns (equal to m) and is written as,

 a11 a1 … a1m 
a 
… a2m 
2
 a21
22
 ⋯ % 
⋯
 am 1 am amm 

2

Definition:

The identity matrix (of order m) is the m  m matrix all of whose

entries are 0 apart from those on the main diagonal, where they all
equal 1,

M Patel (April 2 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

1 0 … 0
def
0 1 … 0 
Im
⋯ ⋯ % 
0 1 
 0

Definition:

The zero matrix (of order m  n) is the m  n matrix all of

whose entries are 0,

0 0 0 … 0
 0 0 0 … 0 
def
Om   0 0 0 … 0
 ⋯ ⋯ % 
n ⋯ 
 0 0 0 0 

For a square zero matrix, sometimes the notation Om is

used.
Matrix Algebra

Addition, Subtraction and Scalar Multiplication

Definition:

The matrix sum of A and B is obtained by adding corresponding

entries of A and B,

def
(a  b )ij aij  bi
j

The matrix difference of A and B is obtained by subtracting

the entries of B from the corresponding ones in A,

def
(a  b )ij aij  bi
j

Note that matrix addition and subtraction only makes sense if A

and B
both have the same order.

M Patel (April 3 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
Example
1 
Add
0 the matrices A = 3  13 1
and
= B
.
   
  4 5
5

As the matrices have the same order, they can be added.

A 0
 B = 3   13 1 
   
 5   4 5 

 50  13 3  1 
=   (4) 2  5
 

13 2
=  
 1 

Example 2
4
Find the difference P  Q where P = 2 0 
 and

 
 4 1 0 37
Q =  2 2 
 .
 8 3 6
P 4 2 0 
 Q =  4 2 2

   8
1 0 37   3 6

 8 0 2
= 9 3 31 
 

Definition:

The scalar multiplication of A by k (k  R ) is obtained by

multiplying each entry of A by k,

def
(ka )ij kai
j

M Patel (April 4 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
Example
3
 4 4 0 
If A = 5 6 1 , calculate 1 A.
  2
8 9 1 
 

1
 4 4 0 
A = 1
5 6 1
2 2  
8 9 1 
 

4 / 2 4 / 2 0/2
 
= 5/2 6/2 1 / 2
8 / 2 9/2 1 / 2
 

2 2 0 
5 
= 3 1
2 2 
4 9
 1
 2 2 

Matrix Multiplication

Definition:

The matrix product of A and B, where A is of order m  n and B

is of order n  p is obtained by the following prescription,

∑ ab
def
(ab)ij ik kj
k  1

(1  i  m and 1  j 
p)
Note that the number of columns of A must equal the number of rows
of
B. To see how to use the horrible prescription, split up A into rows
and B into columns; then taking the ‘ scalar product ’ of the i th row
of A with the j th column of B gives the (i, j ) th entry of AB.

M Patel (April 5 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
Example 4

 1 2 0   3 0 
Calculate AB when A = 8 0 1 and B = 1 2 .
   
 2 4 6   3 4 

As the number of columns of A equals the number of rows of B, the

product AB makes sense.

AB  1 2 0   3 0 
= 8 0 1 1 2
   
 2 4 6   3 4 

 1  3  (2)  1  0  (3) 1  0  (2)  2  0  4

= 
8  3  0  1  1  (3) 8  0  0  2  1  4
 
 (2)  3  4  1  6  (3) (2)  0  4  2  6  4



=  1 4 
21 4
 
 20 32 

In general, AB is not the same matrix as BA.

Example 5
 0 3  1 0
Show that AB ≠ BA for the matrices A = and B = .
   
5  0 

First AB,
AB 0
= 3
1
0
   
5   0 

 0 6 
AB =  
5 

Next BA,

M Patel (April 6 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
1 0 0 3
BA =   
0  5 

0 3
BA =  

10 4

Definition:

When forming AB, B is pre-multiplied by A and A is post-multiplied

by
B.
Definition:
T
The transpose of A (with the order of A being m  n), denoted by A
(sometimes A′ ), is the n  m matrix obtained by interchanging the
rows and columns of A,

def
(aT )ij aji

Example 6
Find 
3 the transpose of P = 7 0
 .
 
5 9 46 

3 5 

P
T
=  7
0 9 
46
 

Definition:

A (square) matrix A can be multiplied by itself any number of

times, giving the n th power of A,

def
n  A _
A
_A__A__ _…

M Patel (April 7 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
Basic Properties of Matrices
 A  B = B  A

 (A  B)  C = A  (B  C )

 k (A  B ) = kA  kB
T T
 (A  B) = A  B T

T
 (A )T = A

 (kA )T = kA T

 A ( BC ) = (AB ) C

 A (B  C ) = AB  AC

 (AB )T = B TA T

 A m
A m = A n
A m
n = A
n

There are 3 important properties that are worth singling out

separately.

 A  O = O  A = A

 AI = AI = A

 AO = AO = O

Thus, the identity and zero matrices behave like the numbers 1
and 0 respectively in ordinary arithmetic and algebra.

Example 7

2 5
For the matrix A 2 qI2 , stating the
=   , show that A = pA
1 16 

values in the form gA hI2 ,
of the integers p and q. Hence write
3
A 
stating the values of g and h.

M Patel (April 8 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

= 2 5  2 5
2
A
 1  
16   1

16 

= 9 90 
2
A
 
18 261

Next,

2 5 1 0 
pA qI2 =  
1 16   q  
 p 0 

pA 
qI 2 =  2p  q 5p 
p 16p  q
 

Equating the 2 matrices gives,

 2p  q 5p  9 90  
p 16p  q = 18 261
   

This immediately gives p = 18; substituting this into one of the

other equations involving q gives q = 27. Thus,
2
A = 18A  27 I2

For the second part,

3 2
A = A A

= (18A  27 I2 ) A

2
= 18 A  27A

= 18 (18A  27 I2 )  27A

= 324A  486 I2  27A

= 297A  486 I2

Hence, g = 297 and h = 486.

M Patel (April 9 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

Example 8

Show that (kABC = k

)T T B TA T .
C

(kABC )T = k (ABC )T

= k ((AB )C )T

= kC T
(AB )T

T T
= kC T
(B A )

= kC T
B TA T

Make sure you can justify each equality in Example 8.

One matrix property that has no counterpart in ordinary arithmetic

and algebra is the fact that the product of 2 matrices can be zero
without either of the matrices being the zero matrix.

Example 9

Calculate AB for the matrices A 1 0 

=   and B  0 0 
 2  = 3 4 .

AB =  1 0   0 0 
   
2  3 

=  0  0 0  0
0  0 0 
0
 

0 0 
=  
0 

M Patel (April 10 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

Special Types of Matrices

Symmetric and Skew-Symmetric Matrices

Definition:

A matrix A is symmetric if,

T
A = A

Note that a symmetric matrix must be square.

Example 10

1 2 4
Show that the matrix A  
 2 0 17 is symmetric.
=

 4 17 6 

T
 1 2 4
T 
A = 2 0 17 

 4 17 6 

Interchanging the rows and columns of A gives,

1 2 4
A
T 
= 2 
 0 17 
 4 17 6 

T
A = A

T
Hence, as A = A, A is symmetric.

Definition:

A matrix A is skew-symmetric (aka anti-symmetric) if,

T
A = A

M Patel (April 11 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
Note that a skew-symmetric matrix must be square and have all
diagonal entries equal to 0.

Example 11

Show that the sum of 2 skew-symmetric matrices is skew-

symmetric. Let the 2 skew-symmetric matrices be A and B. Then,

T T
(A  B) = A  B T

(A  B) T = (A  ( B )
)

(A  B ) T = (A  B )
Hence, as the sum A  B satisfies the skew-symmetric condition,
the sum of 2 skew-symmetric matrices is skew-symmetric.

Orthogonal Matrices

Definition:

A square matrix A (of order n  n ) is orthogonal if,

T
A A = In

Example 12

Show
= that R
 cos θ sin θ  is an orthogonal matrix.
 
  sin θ cos θ 

sin θ 
R T R = cos θ  sin θ   cos θ 
sin θ cos θ   sin θ cos θ
  

 cos
2
 sin θ cos θ 
2
=  sin cos θ sin
θ θ θ
 sin θ cos θ  cos sin sin
2 2
 cos θ 
θ θ θ
 

1 0 
=  
0 

M Patel (April 12 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

= I2

Hence, as R T
R equals the identity matrix, R is orthogonal.

Determinants
The solution to the 1 × 1 system,

ax = b

is,

b
x =
a

assuming a ≠
0.

The solution to the 2 × 2 system of equations,

ax  by 
e cx 
dy  f

is,

de − bf
x = ad − bc

af − ce
y = ad − bc

assuming ad  bc ≠ 0.

The solution to the 3 × 3 system of equations,

ax  by  cz  j
dx  ey  fz  k
gx  hy  iz  l

is (this takes a lot more effort),

M Patel (April 13 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

j (ei − fh ) − k (bi − ch ) + l (bf − ce )

x = a (ei − fh ) − b(di − fg ) + c (dh − eg )

j (fg − di ) − k (cg − ai ) + l (cd − af )

y = a (ei − fh ) − b(di − fg ) + c (dh − eg )

j (dh − ge ) − k (ah − bg ) + l (ae − bd )

z = a (ei − fh ) − b(di − fg ) + c (dh − eg )

assuming a — fh — — fg + c — eg ≠ 0.
(ei ) b(di ) (dh )

In each of these solutions, we require the denominators to be non-

zero. The denominators that arise in the solutions have a pattern
(not necessarily that obvious !) and a special name. We first need
some definitions.

Definition:

A permutation, denoted by σ , of an ordered set of numbers (1, ,

2, 3,
n) is a rearrangement of those numbers.
An even permutation is one where the rearrangement involves an
even number of consecutive switches starting from the original
numbers.

An odd permutation is one where the rearrangement involves an

odd number of consecutive switches starting from the original

Definition:

The sign of a permutation σ , denoted by sign σ , is defined to be

+1
for an even permutation and − 1 for an odd permutation.

Example 13

For the case n = 3, the permutation (1, 2, 3) σ→ (2, 1, 3) is odd, as

the result of σ involves only 1 switch (1 ↔ 2) between
consecutive numbers of (1, 2, 3). The sign of this permutation is − 1
.

M Patel (April 14 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
Example 14

Again, for the case n = 3, the permutation (1, 2, 3) σ→ (3, 1, 2) is

even, as the result of σ involves 4 switches between consecutive
numbers of (1, 2, 3). The switches are (1, 2, 3) → (2, 1, 3) → (2, 3,
1) → (3, 2, 1)
→ (3, 1, 2). For this permutation, sign σ = + 1 .

A permutation can be thought of as a function. For example, in

Example , 1 gets sent to 3 would be written as σ (1) = 3.

Definition:

The determinant of an n  n matrix is given by the Leibniz

formula,
   n 
∑ 
def
det (A )  A  (sign σ   ai σ (i 
  σ )   
   i 1 ) 

Fortunately, this cryptic formula does not need to be remembered;

an alternative formula that is more practical will be given below.

Definition:

For an n  n matrix A, the minor of entry aij is the

(denoted
determinantMij ) of the   (n  1) matrix formed from A
n
(deleting the i row and j column by
th th

of A.

Definition:

The cofactor of entry aij is the

quantity
def
Cij (  1)i j
ij

M
It can be shown that the Leibniz formula for determinants can be
written using cofactors in the form given in the following theorem.

M Patel (April 15 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
Theorem:

The determinant of an n  n matrix is given by the Laplace

expansion formula,

det (A )
= ∑a
j 
ij
(i = 1, 2, 3,
n)
,
1
Cij
The Laplace expansion formula expresses the determinant of a
matrix in terms of smaller determinants. For satisfaction and
reassurance, the following theorems should be proven using the
Laplace expansion formula.

Theorem:

The determinant of a 1  1 matrix is,

A  (a ) = a

Theorem:

The determinant of a 2  2 matrix

is,
a b
A   = ad  bc
 c d

Theorem:

The determinant of a 3  3 matrix

is,

a b c e f  d f  d e
A  d e f  = a    b g  c g
h i   i  

h
 g i 
h

Notice that these are precisely the expressions for the denominators
for the systems at the start of this section.

M Patel (April 16 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
Example 15

Calculate
1 the determinant of the matrix B = 4
2  .
 
 
1
det (B ) = 4 
 
2 

= 1.6  ( 4).2

= 14

Example 16

 1 2 7 
Calculate the determinant of the matrix F = 6 0 1 .
 
3 10 4
 

 1 2 7
det (F ) = 6 0 1 
 3 10 4

 
 0 6
= 1

1 

 ( 2)

1 
  7  6 0 

 10 4  3 4  3 10 

= 1 (0  10)  2 (24  3)  7 ( 60  0)

=  388

Example 17
 2 4 
Solve the equation  = 21 for x.
3x 5
 

 2 4  = 21
3x 5
 

10  12x = 21

M Patel (April 17 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

12x = 11

11
x = 12

Properties of
Determinants

The following properties hold for n × n matrices and k  R .

 det (AB ) = det (A ) det (B )

 det (kA ) = k n
det (A )

T
 det ( A ) = det (A )

Inverse of a Matrix

Inverse Matrices

Of all the operations that have been described for matrices, that of
division has not been mentioned. The closest concept of ‘ division of
matrices ’ involves the following.

Definition:

An n  n matrix A has an inverse if there is a matrix (denoted A

1
) such that,

=
AA 1
A 1 A = In

If a matrix A has an inverse, then A is said to be invertible (aka

non- singular).

If a matrix does not have an inverse, then it is non-invertible

(aka
singular).
Theorem:

A matrix has only 1 inverse.

M Patel (April 18 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

Definition:

The cofactor matrix of a square matrix A is the matrix C whose (i, j

)th entry is Cij .
Definition:

The adjugate (aka classical adjoint) of a square matrix A is the

transpose of the cofactor matrix C,

def
adj (A ) C T

Theorem:

The inverse of a matrix A is given

by,
adj (A )
=
A 1
det (A )

Theorem:

A matrix is invertible iff det (A )  0.

This theorem implies that a matrix is non-invertible iff det (A ) = 0.

Example 18

1 0 4
Show that A  
 2 0 17 is invertible.
=

 4 17 6 

1 0 4
det (A )  
 2 0 17 
=
 4 17 6 

 0 17   2 17   2 0
= 1 
 17 6   0    4  
  4  4 17

M Patel (April 19 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
= 1 (0  289)  4 ( 34  0)

=  425

Hence, as det (A )  0, A is invertible.

Example 19 2 
Determine is singular.
k the values of k for which the matrix
 
 2 k  5

For singularity, we require the determinant of the given matrix to be

0.

k 2 
  = 0
 2 k  5

k (k  5)  ( 2). ( 2) = 0

k 2
 5k  4 = 0

(k  1) (k  4) = 0

Hence, the given matrix is singular when k =  1 and k = 4.

Example 20

2
If the matrix A satisfies the equation A = 18A  27 I2 , show
(without explicitly calculating A 1 ) that A 1 = DA  EI , stating the
2

values of D and E.

2
A = 18A  27 I2

Multiplying (doesn’t matter whether post or pre, as the only

matrices
involved are A and 2I ) this equation throughout by A 1 gives,

1 2 1
A A = A (18A  27 I )
2

Performing the multiplications and simplifying gives,

A 1
= 18 I2  27 A

M Patel (April 20 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

Rearranging gives,
and solving for
1
A 1 2
=  A  I
1
A
2
27 3
Hence, 1
D = 
27 and E = 2 .
3

Inverse and Systems of Equations

An n × n system of equations with coefficient matrix A and solution

represented by the column matrix x can be written as,

Ax = b

where b is the RHS of the system of equations. Assuming the inverse

A 1 exists, multiplying the above equation by A 1 gives,
1
A Ax =
A 1
b
x =
A 1
b

In other words, the following theorem holds.

Theorem:

The n × n system of equations,

Ax = b

1
has a solution if A exists; the solution is then obtained by
calculating,

x =
A 1
b

This result is stated differently in terms of determinants in the

following theorem.

M Patel (April 21 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

Corollary:

A system of n equations in n unknowns has a solution if the

determinant of the coefficient matrix is non-zero.

Example 21

Determine the value of k for which the following system of

equations has a solution.
x
 y  z  1
2x  ky   0

3z
x  4y  z  3

The coefficient matrix is,  

2 k 3
 
1 4 1 
 

The determinant of the coefficient matrix is,

1 1 1 
  k 3 2 3 2 k
2 k 3 = 1    1    1  
1  1  1 1  1 4
4 1  4
 

= (  k  12)  (  2  3)  (8  k)

= 2k  15

According to the previous theorem, the system of equations has a

15
solution when 2k  15  0, i.e. when k  − .
2

In practice, the formula for the inverse involving the adjugate is

difficult to use from scratch. Fortunately, for the cases we are
interested in, there are other approaches.

M Patel (April 22 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

Inverse of a 2 x 2 Matrix

Theorem
:
a b 
The inverse of the 2 × 2 matrix  is,
 c d 

1 d b
1 =  
A a  bc  c a 
d
This is just the adjugate formula for a 2 × 2 matrix, but it is easier
to remember in this form.

Example 22
Find 
7 the inverse of the matrix 3
 .
 
4 

Using the formula in the above theorem gives,

1
7 3
1 2 3
  =  
4  7.2  (  3).4 4 7

1
7 3
1 2 3
  =  
4  26 4 7

It is ok to leave the answer in this form, instead of actually

performing the scalar multiplication; there is no virtue in
complicating an already ugly answer (technically, the answer in the
form above is fully simplified, as
the 1
has been factorised out).
26

Example 23

Solve the following equation for x and y.

M Patel (April 23 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
2 1  x    
    =   8 
16
 2 3 y 
1
2 1
1 3 1
Multiplying this equation  2 , which  (check !),
 3  
8 2 2 
equals
by gives,
 
x  = 1 3 1  8 
    
  y  8 2 2  16

x  3 1 1

  =   
y  2 2  2 

Performing the matrix multiplication gives x = 1 and y = 6. Check

that these values work in the original equation.

Inverse of a 3 x 3 Matrix

Recall that the n × n system A x = b can be solved by calculating

x =
     
 1   0   0
0 1
1 b. Picking b to be =  ,b 2 =  , , b 0 shows that the
A     =   
b1 ⋯ n⋯ ⋯
     
0  0    n 
solution vector x picks out the columns of A 1 ( b picks out the
first
1
1
column of A etc.). Alternatively, the solution vectors x can be
obtained
by row-reducing the augmented matrices,

(A b1 )

(A b2 )

(A bn )

M Patel (April 24 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

so that the LHS of each gives the identity matrix. According to the
first way of obtaining the RHS, what is left on each RHS will be the
columns of A 1
.

These n calculations can be performed in one go by forming a

giant
Augmented matrix consisting of A on
the LHS, but now b1 , b2 , , bn
written beside each other so that the RHS is effectively In .

Theorem:

The inverse of A can be found by row-reducing the extended

Augmented matrix,

(A I )
n

into,

(I n
B)

1
Then B = A .

This theorem is mainly used for a 3 × 3 matrix

A. Example 24

1 0 4
 
Find the inverse of N =  2 0 7  .
 4 1 6 

First set up the big-daddy Augmented matrix,

1 0 1 0 0 
 4

 2 0 7 0 1 0
4 1 6 0 0 1 


Then use EROs to reduce this matrix so that the LHS I3 ; the
becomes RHS will then be N 1 .
M Patel (April 25 St. Machar
2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

R2  
R2 2R1
R3  1 0 2 1 0 0
4R1  

R3 0 0 2 1 0
11 

0 1 2 4 0 1

1
R 
R
2
11
2 1 0 1 0 0
2
 
0 0 1
2 1
11 11 0
 0 1 2 4 0 1 


R2 
R3
1 0 2 1 0 0

0 1 2 4 0 1 

0 0 2 1 
 1 11 0
1 1

R2  
R2 2R3
1 0 2 1 0 0
 
 4011 2
0 1 0 11 1
0 0 2 1 0 
 1 11
1 1

2R3
R1  
R1 7
11  211 0 
1 0 0
 1
0 1  4011 2
11 
0 2
1
1
11 0

0 0 1

1

To get the inverse into a slightly slicker form, factorise out the
fraction,

1  7 2 0
 
M Patel (April 26 St. Machar
2012) Academy
Advanced Higher Notes (Unit Matrice
3) 1 = 40 2 11  s
N 11  2 1 0
 

M Patel (April 27 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
Properties of Inverse Matrices

( 1
 AB ) 1 = B A 1

( T T
 A 1 ) = (A ) 1

det 1
1
 (A ) =
det (A )

 (kA )
1 =
1 A 1

Transformation Matrices

Definition:

A (geometrical) transformation is a way of changing points in space.

In this course, we will focus on transformations in the xy - plane.

A transformation is described by a function.

Definition:

A linear transformation in the plane is a function that sends a point P

( x,
y) to a point Q (ax  by, cx  dy) (a, b, c, d  R ).
Geometrical transformations can be described using matrices.

Theorem:

A linear transformation in the plane can be described as a

matrix equation,

 x'  def  ax  by  a b  x 
     =   
 y'   cx  dy  c d  y 

a b
The matrix   is called the transformation
matrix. c d

M Patel (April 28 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
A transformation matrix can be determined by looking at where
the geometrical transformation sends the points (0, 1) and (1, 0).

Example 25

Find the matrix associated with the transformation (x, y) → (x ’, y

′) where x ′ = 2x  5y and y ′ = x  3y.

Thinking in terms of matrices and vectors,

 x'   a b  x 
  =   
y'
  c d   
y
shows that x ′ = ax  by and y ′ = cx  dy. Comparing this with x ′
=
2x  5y and y ′ = x  3y gives a = 2, b = 5, c = 1, and d = 3.
So, 2 5 
the required transformation matrix is .
1 3 
 

Example 26

Find the image of the point P (3, 1) under the transformation
associated  1  2
with the matrix .
 
3 

The image of P is given by P ′ (x ’, y ′) where,

 x'   1 2  3 
  =   
 y'  3 0   1 

This gives x ′ = 5 and y ′ = 9, Thus, the image of P is P ′ (5, 9).

Invariant Points

Some points are left unchanged under a transformation. If x is a

coordinate vector of a point and T the transformation matrix, then
the following definition can be made.

M Patel (April 29 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
Definition:

An invariant point is a point whose coordinates stay the same

after a transformation,

Tx = x

Example 27

Find the invariant points under the transformation described by

 1 2
the matrix
  .
2 3 

The matrix
equation,
x   1 2   x 
  = 
 y 2 3 
  
y
yields the equations x =  x  2y and y = 2x  3y, which
simplify to x = y and x =  y. Solving these gives x = 0, and y = 0.
Hence, (0, 0) is the only invariant point under this transformation.

There are some standard transformations that must be known.

Reflection in the x - axis

Theorem:

A reflection in the x-axis is described by the transformation

matrix,
1 0
 
 0 1 

Reflection in the y – axis

Theorem:

A reflection in the y-axis is described by the transformation matrix,

M Patel (April 30 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

 1 0
 
0 1

Reflection in the Line y = x

Theorem:

A reflection in the line y = x is described by the transformation

matrix,
0 1
 
1 0

Anticlockwise rotation about the origin

Theorem:

An anticlockwise rotation of angle θ about the origin is described by

the transformation matrix,

cos θ  sin θ 
 cos 
 sin θ θ 

Dilatation (Scaling)

Definition:

A dilatation (aka scaling or homothety) is a transformation that

scales each coordinate of a point by the same amount.

Theorem:

A dilatation is described by the transformation

matrix,
k 0
  (k  R
0 k
)

M Patel (April 31 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s
If k > 1, the dilatation is an enlargement, whereas if k < 1,
the dilatation is a reduction. If k < 0, then the dilatation inverts

Transformation matrices are usually combined to yield a resultant

transformation.

Example 28

Find the transformation matrix associated with a reflection in the x-

axis followed by anti-clockwise rotation of 90° about the origin.

The first transformation is the reflection, which we will label as TR ,

and the second is the rotation, labelled as Tθ . Taking a generic point
P, the first transformation will be TR (P) and the second one will be Tθ
(TR (P)). In other words, we have to perform matrix multiplication
(and remember that, in general, the order of multiplication matters).
So, the required transformation is,

0  1   1 0
Tθ T R =  1 0 0 1 
   

0 1 
=  
1 

Notice that this is the same as reflecting in the line y = x.

Example 29

Find the image of the point P (1, 2) after a reflection in the line y = x
followed by the
dilatation  7 0 
0 7 .

The image of P will be given by,

7 0  0 1  1 
     
0  1   2

M Patel (April 32 St. Machar

2012) Academy
Advanced Higher Notes (Unit Matrice
3) s

Performing this multiplication gives the 14 

  . So, the image of P is
vector P ′ (14, 7). 7 

Transformation of Loci

It is important to know where sets of points, for example, those on

a curve, get mapped to under a transformation.

Example 30

Find the equation of the image of the curve with equation y = x 2

 1 2
under the transformation with associated matrix
 .
2 3 

Let the image of a point P (x, y) be P ′ (x ’, y ′ ). Then,

 x'  1 2 x 
  =    
 y'  2   y

x  1

   1 2  x' 
= 
 y   
2   y' 
x 
   3 2   x'  
y = 2 1 y'
     

  x   3x'  2y'  
y = 2x'  y'
   

Substituting the above expressions for x and y into y = 3 x 2 gives

the
image curve equation in implicit form as 2x'  y' = 3( 3x'  2y' ) 2 .

M Patel (April 33 St. Machar

2012) Academy