Advanced Higher Notes (Unit 3) Matrices

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 a12 a13 … a1n 

 
def  a21 a22 a23 … a2n 
A ≡ ( aij ) m ×n = a a32 a33 … a3n 
 31 
     
a am 2 am 3 amn 
 m1

The element in row i and column j of a matrix is written as aij and called
th
the (i, j) entry of A.

In this course, we will deal almost exclusively with matrices that have
orders 2 × 2 and 3 × 3.

M Patel (April 2012) 1 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

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,

 a11 
 
 a21 
  
 
 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 a12 … a1m 

 
 a21 a22 … a2m 
    
 
a am 2 amm 
 m1

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 2012) 2 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

 1 0 … 0
 
0 1 … 0
def
Im =
   
 
0 0 1 

Definition:

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

entries are 0,

0 0 0 … 0
 
def 0 0 0 … 0
Om × n = 0 0 0 … 0
 
    
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 + bij

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 − bij

Note that matrix addition and subtraction only makes sense if A and B
both have the same order.

M Patel (April 2012) 3 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

Example 1

 0 −3   13 1 
Add the matrices A =   and B =  .
5 2   −4 5 

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

 0 −3   13 1 
A + B =   +  
5 2   −4 5 

 0 + 13 −3 + 1 
=  
 5 + ( −4) 2 + 5 

 13 −2 
=  
1 7 

Example 2

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

 4 −2 0   −4 −2 2 
P − Q =   −  
 1 0 37   −8 3 6 

 8 0 −2 
=  
 9 −3 31 

Definition:

The scalar multiplication of A by k (k ∈  ) is obtained by multiplying

each entry of A by k,

def
(ka )ij = kaij

M Patel (April 2012) 4 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

Example 3

 4 −4 0 
 
If A =  5 6 1  , calculate 1
2
A.
 8 9 −1 
 

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

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

 2 −2 0 
 
=  52 3 1
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 2012) 5 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

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.

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

First AB,

 0 −3   1 0 
AB =    
5 2  0 2

 0 −6 
AB =  
5 4 

Next BA,

M Patel (April 2012) 6 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

1 0   0 −3 
BA =    
0 2 5 2 

 0 −3 
BA =  
 10 4 

Definition:

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

B.

Definition:

The transpose of A (with the order of A being m × n), denoted by A T

(sometimes A′ ), is the n × m matrix obtained by interchanging the rows
and columns of A,

def
(aT )ij = a ji

Example 6

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

 3 5 
T  
P =  −7 9 
 0 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 × … × A

n times

M Patel (April 2012) 7 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

Basic Properties of Matrices

• A + B = B + A

• (A + B) + C = A + (B + C)

• k (A + B ) = kA + kB

T
• (A + B )T = AT + B

• (A T )T = A

• (kA )T = kA T

• A ( BC ) = ( AB ) C

• A (B + C ) = AB + AC

• (AB )T = B TA T

m n
• A A = A m +n = A n
A m

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

M Patel (April 2012) 8 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

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

2
 9 90 
A =  
 18 261 

Next,

2 5  1 0
pA + qI 2 = p   + q  
 1 16  0 1 

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

Equating the 2 matrices gives,

 2p + q 5p   9 90 
  =  
 p 16 p + 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 I 2

For the second part,

3
A = A 2A

= (18A − 27 I 2 ) A

2
= 18 A − 27A

= 18 (18A − 27 I 2 ) − 27A

= 324A − 486 I 2 − 27A

= 297A − 486 I 2

Hence, g = 297 and h = − 486.

M Patel (April 2012) 9 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

Example 8

Show that (kABC )T = k C T B TA T .

(kABC )T = k (ABC )T

= k ((AB )C )T

T
= kC (AB )T

T
= kC (B T A T )

T
= kC 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

1 0 0 0
Calculate AB for the matrices A =   and B =  .
 2 0   3 4 

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

0 + 0 0 + 0
=  
0 + 0 0 + 0

0 0
=  
0 0

M Patel (April 2012) 10 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

Special Types of Matrices

Symmetric and Skew-Symmetric Matrices

Definition:

A matrix A is symmetric if,

AT = 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 
 
AT =  −2 0 17 
 4 17 6 
 

Interchanging the rows and columns of A gives,

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

AT = A

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

Definition:

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

A T = −A

M Patel (April 2012) 11 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

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
(A + B )T = AT + B

(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,

A T A = In

Example 12

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

 cos θ − sin θ   cos θ sin θ 

R TR =    
 sin θ cos θ   − sin θ cos θ 

 cos2 θ + sin2 θ cos θ sin θ − sin θ cos θ 

=  
 sin θ cos θ − cos θ sin θ sin2 θ + cos2 θ
 

 1 0
=  
0 1 

M Patel (April 2012) 12 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

= 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 2012) 13 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

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 (ei − fh ) − b (di − fg ) + c (dh − eg ) ≠ 0.

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 numbers.

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

σ
For the case n = 3, the permutation (1, 2, 3)  → (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 2012) 14 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

Example 14

σ
Again, for the case n = 3, the permutation (1, 2, 3)  → (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 = 

 σ
(sign σ ) 





∏
i =1
ai σ (i )




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 determinant

(denoted Mij ) of the (n − 1) × (n − 1) matrix formed from A by
deleting the i th row and j th column of A.

Definition:

The cofactor of entry aij is the quantity

def
Cij = ( − 1)i + j Mij

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 2012) 15 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

Theorem:

The determinant of an n × n matrix is given by the Laplace expansion

formula,

det (A ) = ∑a C
j =1
ij ij
(i = 1, 2, 3, … , n)

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   + c  
g h i  h i  g i  g h


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

M Patel (April 2012) 16 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

Example 15

 1 −4 
Calculate the determinant of the matrix B =  .
2 6 

 1 −4 
det (B ) =  
2 6 

= 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 −1   6 −1  6 0 
= 1   − ( − 2)   + 7  
 −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 2012) 17 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

12x = 11

11
x =
12

Properties of Determinants

The following properties hold for n × n matrices and k ∈  .

• det (AB ) = det (A ) det (B )

n
• det (kA ) = k det (A )

• det ( A T ) = 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:

−1
An n × n matrix A has an inverse if there is a matrix (denoted A )
such that,

A A −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 2012) 18 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

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
T
adj (A ) = C

Theorem:

The inverse of a matrix A is given by,

−1 adj (A )
A =
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   − 0   + 4  
 17 6  4 6  4 17 

M Patel (April 2012) 19 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

= 1 (0 − 289) + 4 ( − 34 − 0)

= − 425

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

Example 19
k −2 
Determine the values of k for which the matrix   is singular.
 −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 I 2 , show
−1 −1
(without explicitly calculating A ) that A = DA + EI2 , stating the
values of D and E.

2
A = 18A − 27 I 2

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

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

−1 2 −1
A A = A (18A − 27 I 2 )

Performing the multiplications and simplifying gives,

−1
A = 18 I2 − 27 A

M Patel (April 2012) 20 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

−1
Rearranging and solving for A gives,

−1 1 2
A = − A + I
27 3 2

1 2
Hence, D = − and E = .
27 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 −1
A Ax = A b

−1
x = A 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,

−1
x = A b

This result is stated differently in terms of determinants in the following

theorem.

M Patel (April 2012) 21 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

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 + 3z = 0
x + 4y − z = 3

The coefficient matrix is,

1 1 1 
 
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 4 −1   4 −1   1 −1  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 2012) 22 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

Inverse of a 2 x 2 Matrix

Theorem:

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

−1 1  d − b
A =  
ad − bc  − c a 

This is just the adjugate formula for a 2 × 2 matrix, but it is easier to

remember in this form.

Example 22

 7 −3 
Find the inverse of the matrix  .
4 2 

Using the formula in the above theorem gives,

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

−1
 7 −3  1  2 3
  =  
4 2  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
1
the has been factorised out).
26

Example 23

Solve the following equation for x and y.

M Patel (April 2012) 23 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

 2 1  x   8 
 
   
  = 16
 −2 3   y 


−1
 2 1 1  3 −1 
Multiplying this equation by   , which equals   (check !),
 −2 3  8 2 2 
gives,

  = 1  3 −1   8 
x   

 y    
8  2 2  16

  =  3 −1   1 
x   

 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   0
A b. Picking b to be b1 =   , b2 =   , … , bn =   shows that the
−1  
  

        
     
0 0 n 
−1
solution vector x picks out the columns of A ( b1 picks out the first
column of A −1 etc.). Alternatively, the solution vectors x can be obtained
by row-reducing the augmented matrices,

(A b1 )

(A b2 )

(A bn )

M Patel (April 2012) 24 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

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 In )

into,

(In
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 4 1 0 0
 
 −2 0 7 0 1 0
 4 1 6 0 0 1 


Then use EROs to reduce this matrix so that the LHS becomes I3 ; the
−1
RHS will then be N .

M Patel (April 2012) 25 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

R2 → R2 + 2R1
R3 → R3 − 4R1
1 0 2 0 0
1
 
 0 0 11 2 1 0
 0 1 −2 −4 0 1 


1
R2 ↔ R
11 2
1 0 2 1 0 0
 
0 0 1 0
2 1
11 11
 0 1 −2 −4 0 1 


R2 ↔ R3
1 0 2 1 0 0
 
 0 1 −2 −4 0 1
0 0 1 2 1 0 
 11 11

R2 → R2 + 2R3
1 0 2 1 0 0
 
0 1 0 − 40 11 1
2
11
0 0 1 2 1 0 
 11 11

R1 → R1 − 2R3
1 0 0 7
11 − 2 11 0 
 
0 1 0 − 40 11 1
2
11
0 0 1 2 1 0 
 11 11

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

 7 −2 0 
−1 1  
N =  −40 2 11 
11
 2 1 0 


M Patel (April 2012) 26 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

Properties of Inverse Matrices

−1
• (AB ) −1 = B A −1

• (A −1 ) T = (A T ) −1

−1 1
• det ( A ) =
det (A )

1
• (kA ) −1 = A −1
k

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 ∈  ).

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 2012) 27 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

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 0 

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 2012) 28 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

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 the

 1 2
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 2012) 29 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

 −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 
 
 sin θ cos 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 ∈  )
0 k 

M Patel (April 2012) 30 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

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 TR =    
 1 0   0 −1 

0 1 
=  
 1 0

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
7 0
followed by the dilatation  .
0 7

The image of P will be given by,

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

M Patel (April 2012) 31 St. Machar Academy

Advanced Higher Notes (Unit 3) Matrices

 14 
Performing this multiplication gives the vector   . So, the image of P is
7
P ′ (14, 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 under
 1 2
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 3  y 

−1
x   1 2  x' 
  =    
y  2 3  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 2012) 32 St. Machar Academy