CHAPTER (4)

MATLAB
MATRICES
 1. Defining Matrices

 You can define a matrix by typing in a list of numbers enclosed in square brackets. You
can separate the numbers by commas or by spaces or combine the two techniques in the
same matrix definition. To indicate a new row, you can use a semicolon:

a = [2.5,1.1]; or a = [2.5 1.1]; or a = [2.5, 1.1]; b = [-1, 0, 2; 1, 3, 0; -7, 0, 2];

 If there are too many numbers in a row to fit on one line, you can continue the statement
on the next line, but a comma and an ellipsis (…) are required at the end of the line,
indicating that the row is to be continued.

a=[2, 4, 8, 26, 197, 3, 90, 38, 28, 221, 234, 15, 97, 83, 90];
or
a=[2, 4, 8, 26, 197, 3, 90, 38, 28, ...,
221, 234, 15, 97, 83, 90];

 You can define a matrix in terms of another matrix that has already been defined.
a=[1, 4, 12], b=[2, 6, 9], c=[a; b]
→ c=
1 4 12
2 6 9

 We can change values in a matrix, or include additional values, by using an index

number to specify a particular element.
a=[2, 3, 4, 5]

a(3)=-7: changes the third value in the matrix a from 4 to -7.

a(3)=-7 → a=[2,3,-7,5]

a(5)=12: extends the matrix a by defining a fifth element.

a(5)=12 → a=[2 3 4 5 12]

If we define element a(9) = 1.4; matrix a will have eight values, and the values of a(5),
a(6), a(7), and a(8) will be set to 0
a(9)=1.4 → a=[2, 3, 4, 5, 0, 0, 0, 0, 1.4]

P a g e | 29
2. Extracting data from matrices

a=[2, 3, 4, 5; 8, 7, 6, 5; -3, 5, -2, 1]

x=a(r, :) : extracts row r from matrix a.

x=a(2, :) → a=[8 7 6 5]

x=a(:, c) : extracts column c from matrix a.

x=a(:,2) → a=[ 3; 7; 5]

x=a(r1:r2, :) : extracts from row r1 to row r2 from matrix a.

x=a(2:3, :)
→x =
8 7 6 5
-3 5 -2 1

x=a(:, c1:c2) : extracts from column c1 to column c2 from matrix a.

x=a(:,2:4)
→x =
3 4 5
7 6 5
5 -2 1

x=(r1:r2, c1:c2) extracts from row r1 to r2 and from column c1 to column c2 from a
matrix a.
x=a(1:3, 2:4)
→x =
3 4 5
7 6 5
5 -2 1

P a g e | 30
 3. Special Matrices
 zeros(m): Creates an m×m matrix of zeros.
a=zeros(2) → a = 0 0
0 0

 zeros(m,n): Creates an m×n matrix of zeros.

a=zeros(2,4) →a =
0 0 0 0
0 0 0 0

 ones(m) Creates an m × m matrix of ones.

a=ones(2)
→a=
1 1
1 1

 ones(m,n) Creates an m × n matrix of ones.

a=ones(2,5)
→ a=
1 1 1 1 1
1 1 1 1 1

 eye(m) Creates an m × m Identity matrix

a = eye(3)
→ a=
1 0 0
0 1 0
0 0 1

 eye(m,n) generates an m × n Identity matrix

a = eye(3,5)
→ a=
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0

 diag(a) Extracts the diagonal of a two-dimensional matrix a.

a=[1,4,5; 2,5,7; 8,9,6] b=[2,3,4,5; 7,6,4,8] (not square ?)

c= diag(a) → c =[1; 5; 6] d=diag(b) → d =
2
6 ?

P a g e | 31
  diag(a)For any vector a, creates a square matrix with a as the diagonal.
a=[-2, 4, 7] b=diag(a)

→b =
-2 0 0
0 4 0
0 0 7

 magic(m) Creates an m×m “magic” matrix.

a= magic(3) → a= 8 1 6
3 5 7
4 9 2

4. Some matrix operations and functions

(') transpose operation: changes the rows of a matrix into columns and the columns into
rows

a=[2,7; 5,3;9,7;-3,1]
b=a' → b=
2 5 9 -3
7 3 7 1

(*) matrix multiplication a*b,

note:
- the number of columns of a must be equal to the number of rows of b
- a*b=b*a (in general)

a= [2,-3,5; -9,7,12;1,-1,8],b=[3,2; -4,5;4,1]

c=a*b → c=
38 -6
-7 29
39 5

d=b*b → ??? Error using ==>m times

Inner matrix dimensions must agree.

P a g e | 32
(^) matrix power.

a=[1,3,2;4,1,6;5,6,5], b=[2,-1,2; 3,-2,1; 6,7,9]

a^2=[23 18 30; 38 49 44; 54 51 71]
a^1.5=
6.4311 - 1.0283i 6.5499 + 2.1311i 8.1435 - 1.1446i
11.6708 + 0.8462i 11.8866 - 4.7337i 14.7785 + 3.1220i
15.9524 + 0.1314i 16.2473 + 2.1244i 20.2002 - 1.6072i

b^2=[13 14 21; 6 8 13; 87 43 100]

b^1.5=
2.6502 + 0.0000i 3.5656 - 0.0000i 6.6393
2.7926 + 0.0000i -0.2302 - 0.0000i 4.0222 - 0.0000i
26.6625 + 0.0000i 14.6660 - 0.0000i 29.6815 + 0.0000i

Note:
Raising a matrix c to the matrix power (^)of two is different from raising c to the array
power (.^) of two.

c=[3, 5, 2; 7, 2.4, 1.3; 5.5, 6, 2.8]

c^2=
55.0000 39.0000 18.1000
44.9500 48.5600 20.7600
73.9000 58.7000 26.6400

c.^2=
9.0000 25.0000 4.0000
49.0000 5.7600 1.6900
30.2500 36.0000 7.8400

inv or ^-1 determines the inverse of a square matrix

Note: a*inv(a)=inv(a)*a=eye(a) , the determinant of a must be not equal zero.

a=[1,3,2;4,1,6;5,6,5], b=[2,-1,2; 3,-2,1]

c = inv(a)
→ c=
-0.8378 -0.0811 0.4324
0.2703 -0.1351 0.0541
0.5135 0.2432 -0.2973

OR c = a^-1

d = inv(b)
→ d =??? Error using ==> inv Matrix must be square.

P a g e | 33
 5. Solution of Linear equations
Given:

𝑎11 𝑥 + 𝑎12 𝑦 + 𝑎13 𝑧 = 𝑟1

𝑎21 𝑥 + 𝑎22 𝑦 + 𝑎23 𝑧 = 𝑟2
𝑎31 𝑥 + 𝑎32 𝑦 + 𝑎33 𝑧 = 𝑟3

or in matrix form Ax= r, with

𝑎11 𝑎12 𝑎13 𝑥 𝑟1
𝐀 = [𝑎21 𝑎22 𝑎23 ] , 𝐱 = [𝑦] , 𝐫 = [𝑟2 ]
𝑎31 𝑎32 𝑎33 𝑧 𝑟3

Three methods for solution:

a. Solution Using the Matrix Inverse

x = inv(A)*r

to be avoided since it can result excessive round-off errors

a=[2,3,-2;-1,-1,-1;1,-2,3], r=[-5;-4;13]

x = inv(a)*r → x = [2, -1,3]

b. Solution Using Matrix left Division

x = A\r

a=[2,3,-2;-1,-1,-1;1,-2,3], r=[-5;-4;13]
x = a\r → x=
2.0000
-1.0000
3.0000

P a g e | 34
c. Solution Using the Reduced Row Echelon Function

b = [a,r] (augmented matrix, expanded matrix)

rref(b)

a=[2,3,-2;-1,-1,-1;1,-2,3], r=[-5;-4;13]
b = [2, 3, -2, -5; -1, -1, -1, -4; 1, -2, 3, 13]
rref(b) →
1 0 0 2
0 1 0 -1
0 0 1 3

MATLAB is capable of solving problems which are either overdefined or underdefined

using left division.

a=[3,2,5;4,5,-2;1,1,1;2,-4,-7], r=[22;8;6;-27]
x=a\r →
1.0000
2.0000
3.0000

If the system is underdefined then MATLAB solves the problem by setting some variables
equal to zero.
a=[2,5,3;6,12,8], r=[18;8]
→ ans =
-29.3333
15.3333
0

a=[3,2,5,7;4,5,-2,3], r=[20;12]
→ ans =
0
0.8276
0
2.6207

P a g e | 35