MATLAB TUTORIAL

Necmettin Yildirim

Division of Natural Sciences

New College of Florida

5800 BayShore Road, Sarasota FL 34243

August, 2016

1
 Contents

Chapter 1. WHAT IS MATLAB? 4

1.1. MATLAB at New College 4
1.2. MATLAB Screen 5
1.3. Command Window 5
1.4. Using the Help Browser 5
1.5. Reserved MATLAB Variables 6
1.6. Vectors 6
1.7. What is in the memory of MATLAB 7
1.8. Matrices 8
1.9. Algebra with Matrices in MATLAB 9
1.10. Problems 10

Chapter 2. OPERATIONS with MATRICES in MATLAB 11

2.1. Solving a Linear System of Equations (Bx = t) 13
2.2. Randomly Generated Matrices 14
2.3. floor, round and ceil 15
2.4. Output formats 15
2.5. MATLAB Operations: Summary 17
2.6. Some Useful Built-In Mathematical Functions in Matlab 17
2.7. MATLAB Built-in Vector Functions 17
2.8. A few more useful built-in functions in Matlab 18
2.9. Concatenation of Matrices and Arrays in MATLAB 19
2.10. Problems 22

Chapter 3. FUNCTIONS and SCRIPTS in MATLAB 24

3.1. Inline functions 24
3.2. Saving and Loading your work in an “.m” file 25
3.3. Plot functions 25
3.4. Plot multiple curves in 2D with plot: Vector vs Matrix 27
3.5. Plotting curves in 3D with plot3 28
3.6. Plotting in 3D with plot3 29
3.7. Plotting in 3D with surf 29
2
 CONTENTS 3

3.8. Plotting in 3D with surfc 30

3.9. image 30
3.10. Problems 31

Chapter 4. WRITING YOUR OWN FUNCTIONS and SCRIPTS in MATLAB 32

4.1. Driver codes and functions in MATLAB 32
4.2. if...elseif...else...end conditional statements 34
MATLAB’s Relational and Logical Operators 35
4.3. Loops in MATLAB 37
4.4. Structure of for...end loop 37
4.5. Structure of while...end loop 40
4.6. Problems 44

Chapter 5. SOME BUILT-IN FUNCTIONS in MATLAB 46

5.1. roots: Polynomial roots 46
5.2. fsolve: Solving system of nonlinear equations with several variables 47
5.3. fminbnd Single-variable bounded nonlinear function minimization 48
5.4. fminsearch Multidimensional unconstrained nonlinear minimization 49
5.5. lsqcurvefit: solves non-linear least squares problems 51
5.6. MATLAB: Using the Debugger 51
5.7. MATLAB Hints (Source: Tobin A. Driscoll 2009) 52
5.8. Problems 53

Chapter 6. SOLVING ORDINARY DIFFERENTIAL EQUATIONS in MATLAB 56

6.1. Solving an ODE with an inline function 56
6.2. Solving an ODE with “.m” script and a function 57
6.3. Solving a System ODEs 58
 CHAPTER 1

WHAT IS MATLAB?

• MATLAB is a high level language with many specialized functions and toolboxes to
ease simulation
• MATLAB stands for Matrix Laboratory
• Everything in MATLAB is encoded in a matrix
• This tutorial covers the basics of MATLAB, including matrices and functions, as
well as simulation of ordinary differential equations (ODEs)

1.1. MATLAB at New College

• New College has a MATLAB server. You must login to the server to run MATLAB.
• You need a user account in order to be able to connect to the server.
• To connect to the server, use Windows Remote Desktop Connection(RDC). If you
have a Mac, you can download Microsoft Remote Desktop for free
• In Windows, you can access RDC using Start=>Accessories=>Remote Desktop
Connection. Both PC and Mac users should then type matlab and enter your NCF
email credential.

4
 1.4. USING THE HELP BROWSER 5

1.2. MATLAB Screen

1.3. Command Window

• You can use MATLAB’s command window as a fancy calculator. In the command
window type the followings:

>‌> 2+2 % notice that MATLAB can simply be a calculator

ans =
4
>‌> x=10 % sets x to 10
x =
10
>‌> x=10;% the semicolon suppresses the output
>‌>

• MATLAB ignores anything after the comment sign “%”

• Use % to add comments in your code

1.4. Using the Help Browser

You can always consult the help browser

 1.6. VECTORS 6

• MATLAB has thousands of built-in functions. You can search and get some help if
you know the name of a MATLAB function

>‌> help “name of a function”

Example 1. To get some help on MATLAB’s square root function "sqrt", just type
»help sqrt.

>‌>help sqrt
SQRT Square root.
SQRT(X) is the square root of the elements of X.
Complex results are produced if X is not positive.
See also sqrtm, realsqrt, hypot.
Reference page in Help browser doc sqrt

1.5. Reserved MATLAB Variables

Special variable names:

• ans : the answer of the last unassigned expression

• pi : 3.1415.....
• eps : the smallest possible number on this computer

1.6. Vectors

(A) Row Vectors

>‌> x=[2, 3, 6]
x =
2 3 6
>‌> 2*x % you can multiply a vector by a constant
ans =
4 6 12
>‌> 5+x % you can add a constant to a vector
ans =
7 8 11
>‌> x=[2 3 6] % commas are optional
x =
2 3 6

(B) Column Vectors

 1.7. WHAT IS IN THE MEMORY OF MATLAB 7

>‌> x=[2; 3; 6] % you have to put “;” between numbers

x =
2
3
6
>‌> x’ % transpose of x
ans =
2 3 6
>‌>x=2:2:10 %this notation is the same as x=[2 4 6 8 10]
x =
2 4 6 8 10
>‌> y=sin(x) % creates a new vector of values from x.
y =
0.9093 -0.7568 -0.2794 0.9894 -0.5440

>‌>x=linspace(X1, X2, N)
% generates a row vector of N linearly equally spaced points between X1 and X2.

x=linspace(1,3,5) % generates 5 numbers between 1&3

x=linspace(1,3) % generates 100 numbers between 1& 3

1.7. What is in the memory of MATLAB

• whos : List current variables in the memory

• clear : clear variables and functions memory
• clear all : removes all variables from memory

>‌> whos % List current variables

Name Size Bytes Class Attributes
Example 1x1 976 struct
ans 1x2 16 double
x 1x5 40 double
y 1x5 40 double
>‌> clear x
>‌> whos
Name Size Bytes Class Attributes
Example 1x1 976 struct
ans 1x2 16 double
y 1x5 40 double
>‌>
 1.8. MATRICES 8

1.8. Matrices

>‌> A= [ 1 2; 3 4] % defining a matrix

A =
1 2
3 4
>‌> B= [ 1 1; 1 1];
>‌> A*B % the symbol “*” is matrix multiplication
ans =
3 3
7 7
>‌> A.*B % term by term multiplication
ans =
1 2
3 4

>‌> A^2 % taking square of A, which is A*A

ans =
7 10
15 22
>‌> A.^2 % taking term by term square
ans =
1 4
9 16
>‌> A.^4 % fourth power of each term
ans =
1 16
81 256

1.8.1. Special Matrices.

>‌> eye(5) % create the "identity matrix" of size 5

ans =
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1
>‌> ones(4) % create a square matrix of ones
ans =
 1.9. ALGEBRA WITH MATRICES IN MATLAB 9

1 1 1 1
1 1 1 1
1 1 1 1
1 1 1 1

>‌> 5*eye(4)
ans =
5 0 0 0
0 5 0 0
0 0 5 0
0 0 0 5
>‌> ones(2,4) % create a 2x4 matrix with ones in every entry
ans =
1 1 1 1
1 1 1 1
>‌> zeros(2,4) % create a 2x4 matrix of zeros
ans =
0 0 0 0
0 0 0 0
>‌>

1.9. Algebra with Matrices in MATLAB

>‌> x=[2 4 6]
average = x*ones(3,1)/3
x =
2 4 6
average =
4
>‌>

Exercise 2. Try the same line as above with ones(1,3) replacing ones(3,1). What happens?

• Punctuation: If you want to continue a line, use ... at the end

>‌> longone = 1+2+3+4+5+ ...

6+7+8+9+10
longone =
55

• Most mathematical functions work as you’d expect:

 1.10. PROBLEMS 10

x=[0 1]
sin(x)
asin(x)
exp(x)
log(x) % ?? Log of zero? What’s that?
" #
1 2
Example 3. Apply sqrt to a simple 2 × 2 array of numbers such as .
2 4

1.10. Problems
35.7×64−73 37
(1) Compute 45+52
and 54 − 7 × 62 + 93 −652
2
(2) Evaluate a + abc + (a+0.5(ab−c))
√ if a = 15.62; b = −7.08; c = 62.5.
|ab|
(3) Use MATLAB’s linspace to create the following matrix,
 
0 4 8 12 16 20 24 28
A =  69 68 67 66 65 64 63 62 
 

1.4 1.1 0.8 0.5 0.2 −0.1 −0.4 −0.7

 CHAPTER 2

OPERATIONS with MATRICES in MATLAB

>‌> A = [1 2 3; 3 4 5; 6 7 8]
A =
1 2 3
3 4 5
6 7 8
>‌> A(2,3) % entry located in 2nd row and 3rd column of A
ans =
5
>‌> A(4,1)
??? Index exceeds matrix dimensions.

Submatrices can be “extracted”

>‌> A(1:2,2:3) % extracts first two rows and 2nd and 3rd columns
ans =
2 3
4 5
>‌> A(:,2) % extracts 2nd column
ans =
2
4
7
>‌> A(3,:)% extracts 3nd row
ans =
6 7 8
>‌> A([1 3],[2 3]) % extracts 1st&3rd rows,2nd&3rd columns
ans =
2 3
7 8
>‌> K=[1 2 3 4 5; 6 7 8 9 10; 11 12 13 14 15; 16 17 18 19 20]
K =
11
 2. OPERATIONS WITH MATRICES IN MATLAB 12

1 2 3 4 5
6 7 8 9 10
11 12 13 14 15
16 17 18 19 20
>‌> K(1:2:3,2:3:end) % extract 1st,3rd rows and 2nd&5th columns (until last)
ans =
2 5
12 15

>‌> B=[-1 3 10;-9 5 25;0 14 2];

>‌> s=[-1 8 5];
>‌> t=[7;0;11];
>‌> B-2 % subtracts 2 from each entry in B
ans =
-3 1 8
-11 3 23
-2 12 0
>‌> A+B % sum of A and B
ans =
0 5 13
-6 9 30
6 21 10

>‌> A*B % product of A and B

ans =
-19 55 66
-39 99 140
-69 165 251
>‌> s-t
??? Error using ==> minus
Matrix dimensions must agree.
>‌> s-t’ % subtracts transpose of t from s
ans =
-8 8 -6
>‌> B*t % multiplies B and t
ans =
103
212
22
 2.1. SOLVING A LINEAR SYSTEM OF EQUATIONS (Bx = t) 13

>‌> B*s % tries to multiply B and s

??? Error using ==> mtimes
Inner matrix dimensions must agree.
>‌> B*s’ % multiplies B and transposeof s
ans =
75
174
122

2.1. Solving a Linear System of Equations (Bx = t)

x = B −1 t is the solution of the system of linear equation Bx = t, where x and t are column
vectors, and B is a matrix. Note the important differences between row vectors and column
vectors!

>‌> B
B =
-1 3 10
-9 5 25
0 14 2
>‌> t
t =
7
0
11

>‌> x=B\t % solution of Bx=t

x =
2.4307 0.6801 0.7390

>‌> B
B =
-1 3 10
-9 5 25
0 14 2
>‌> s
s =
-1 8 5
>‌> y=s/B % solution of yB=s
 2.2. RANDOMLY GENERATED MATRICES 14

x =
0.1686 0.0924 0.5023

2.2. Randomly Generated Matrices

• rand : Generates UNIFORMLY distributed pseudorandom numbers on the unit in-

terval (0, 1)
• randn: Generates NORMALLY distributed pseudorandom numbers with mean µ = 0
and standard deviation σ = 1

>‌> rand(4) % 4x4 uniform random matrix from (0,1)

ans =
0.8909 0.1493 0.8143 0.1966
0.9593 0.2575 0.2435 0.2511
0.5472 0.8407 0.9293 0.6160
0.1386 0.2543 0.3500 0.4733
>‌> 2*rand(4) % 4x4 uniform random matrix from (0,2)
ans =
0.7033 1.8344 0.7609 1.0616
1.6617 0.5717 1.1356 1.5583
1.1705 1.5144 0.1517 1.8680
1.0994 1.5075 0.1079 0.2598
>‌> 1+2*rand(4) % 4x4 uniform random matrix from (1,3)
ans =
2.1376 1.3244 1.3313 2.3784
1.9388 2.5886 2.2040 2.4963
1.0238 1.6224 1.5259 1.9011
1.6742 2.0571 2.3082 1.1676
>‌> randn(5) % from normal distribution with 0 mean and 1 std
ans =
-0.8396 1.4367 0.8252 1.0984 -0.8236
1.3546 -1.9609 1.3790 -0.2779 -1.5771
-1.0722 -0.1977 -1.0582 0.7015 0.5080
0.9610 -1.2078 -0.4686 -2.0518 0.2820
0.1240 2.9080 -0.2725 -0.3538 0.0335

Example 4. How can you generate 100 random numbers from the uniform distribution on
the interval [a, b]?

Answer: r = a + (b-a).*rand(100,1);
 2.4. OUTPUT FORMATS 15

Example 5. How can you generate a 6x6 random matrix from the uniform distribution on
the interval [15, 25]?

Answer: 15+round(10*rand(6))

>‌> A= 15+round(10*rand(6))
A =
24 22 24 15 20 16
24 20 23 20 16 23
18 20 21 17 22 23
22 24 17 25 15 22
17 21 17 22 16 16
15 21 24 20 20 22

2.3. floor, round and ceil

• floor : Round towards negative infinity.

• ceil : Round towards positive infinity.
• round : Round towards nearest integer.

>‌> round(-1.8)
ans =
-2
>‌> ceil(-1.8)
ans =
-1
>‌> floor(-1.8)
ans =
-2

2.4. Output formats

• format short/long/short e/long e: Set output formats.

>‌> pi
ans =
3.1416
>‌> format long
>‌> pi
ans =
3.141592653589793
 2.4. OUTPUT FORMATS 16

>‌> format short e

>‌> pi
ans =
3.1416e+000
 2.7. MATLAB BUILT-IN VECTOR FUNCTIONS 17

2.5. MATLAB Operations: Summary

+ : addition
- : subtraction
* : multiplication
^ : power
’ : transpose
\ : left division
/ : right division
.* : term by term multiplication
.^ : term by term power
.\ : term by term left division
./ : right division

2.6. Some Useful Built-In Mathematical Functions in Matlab

sin : trigonometric sine

cos : trigonometric cosine
tan : trigonometric tangent
asin : trigonometric inverse sine (arcsine)
acos : trigonometric inverse cosine (arccosine)
atan : trigonometric inverse tangent (arctangent)
exp : exponential log natural logarithm
abs : absolute value
sqrt : square root
rem : remainder

2.7. MATLAB Built-in Vector Functions

max : largest component

min : smallest component
length : length of a vector
sort : sort in ascending order
sum : sum of elements
prod : product of elements
median : median value (middle value when sorted out )
mean value (x = N1 N
P
mean : i=1 xiq
)
: standard deviation (x = N1−1 N 2
P
std i=1 (xi − x) )
 2.8. A FEW MORE USEFUL BUILT-IN FUNCTIONS IN MATLAB 18

>‌> u=[0.0155 0.9841 0.1672 0.1062 0.3724];

>‌> max(u)
ans =
0.9841
>‌> sort(u)
ans =
0.0155 0.1062 0.1672 0.3724 0.9841

Example 6. max/min/sort... works for matrices column wise

>‌> A=rand(3,4) % generate a random 3x4 matrix from uniform distribution

A =
0.1981 0.9203 0.4228 0.9831
0.4897 0.0527 0.5479 0.3015
0.3395 0.7379 0.9427 0.7011

>‌> max(A)
ans =
0.4897 0.9203 0.9427 0.9831

2.8. A few more useful built-in functions in Matlab

size : size of a matrix

det : determinant of a square matrix
inv : inverse of a matrix
rank : rank of a matrix
rref : row reduced echelon form
eig : eigenvalues and eigenvectors
poly : characteristic polynomial
lu : LU factorization
qr : QR factorization
>‌>A=[-1 2 5;6 -1 4; 1 0 1];
>‌> det(A)
ans =
2.0000
>‌> inv(A)
ans =
-0.5000 -1.0000 6.5000
 2.9. CONCATENATION OF MATRICES AND ARRAYS IN MATLAB 19

-1.0000 -3.0000 17.0000

0.5000 1.0000 -5.5000
>‌> A*inv(A) % should give the identity matrix
ans =
1.0000 0 0.0000
0.0000 1.0000 -0.0000
0.0000 0.0000 1.0000
>‌> [V,D]=eig(A) % computes eigenvalues and eigenvectors of A
V =
0.5718 -0.5363 -0.3319
-0.8143 -0.8225 -0.8947
-0.0999 -0.1893 0.2989
D =
-4.7221 0 0
0 3.8326 0
0 0 -0.1105

Exercise 7. Use MATLAB to check if the columns of V are eigenvectors and the diagonal
entries of D are corresponding eigenvalues of U

>‌> A*V(:,1)-D(1,1)*V(:,1)
ans =
1.0e-014 *
0.1332 -0.0444 -0.0222

2.9. Concatenation of Matrices and Arrays in MATLAB

You can concatenate matrices and arrays in MATLAB. See the examples below.
Let a, b, A and B be the following matrices

>‌> a=rand(2,2)
a =
0.8407 0.8143
0.2543 0.2435
>‌> b=rand(3,2)
b =
0.9293 0.2511
0.3500 0.6160
0.1966 0.4733
 2.9. CONCATENATION OF MATRICES AND ARRAYS IN MATLAB 20

>‌> A=[1:5;6:10]% Define a 2x5 matrix

A =
1 2 3 4 5
6 7 8 9 10
>‌> B=3*[1:5;6:10]% Define a 2x5 matrix
B =
3 6 9 12 15
18 21 24 27 30

>‌> C=[A B] % Create 2x10 matrix

C =
1 2 3 4 5 3 6 9 12 15
6 7 8 9 10 18 21 24 27 30

>‌> D=[b’ a]] % 2x5 matrix

D =
0.9293 0.3500 0.1966 0.8407 0.8143
0.2511 0.6160 0.4733 0.2543 0.2435

>‌> E=[A;B]% 4x5 matrix

E =
1 2 3 4 5
6 7 8 9 10
3 6 9 12 15
18 21 24 27 30

>‌> F=[A;[b’ a]] % 4x5 matrix

F =
1.0000 2.0000 3.0000 4.0000 5.0000
6.0000 7.0000 8.0000 9.0000 10.0000
0.9293 0.3500 0.1966 0.8407 0.8143
0.2511 0.6160 0.4733 0.2543 0.2435

>‌> Z(:,:,1)=A; Create a 3D matrix

>‌> Z(:,:,2)=B;
>‌> Z
 2.9. CONCATENATION OF MATRICES AND ARRAYS IN MATLAB 21

Z(:,:,1) =
1 2 3 4 5
6 7 8 9 10
Z(:,:,2) =
3 6 9 12 15
18 21 24 27 30

>‌> F=Z(1,:,:) % first front slide

F(:,:,1) =
1 2 3 4 5
F(:,:,2) =
3 6 9 12 15
>‌> F2=Z(2,:,:)% second front slide
F2(:,:,1) =
6 7 8 9 10
F2(:,:,2) =
18 21 24 27 30

>‌> G=Z(:,1,:)% first xx slide

G(:,:,1) =
1
6
G(:,:,2) =
3
18
>‌> G4=Z(:,4,:)% 4th xx slide
G4(:,:,1) =
4
9
G4(:,:,2) =
12
27

>‌> A(:,2)=[ ] % Eliminate the second column

A =
1 3 4 5
6 8 9 10
>‌> A(:,[3 4])=[ ]% Eliminate the third and fourth columns
 2.10. PROBLEMS 22

A =
1 3
6 8
>‌> A(1,:)=[ ]% Eliminate the first row
A =
6 8

2.10. Problems
   
2 9 0 0 −1
 0 4 1 4   6  h i
(1) Define A =  , b =  and a = 3 −2 4 −5 and calculate
   
 7 5 5 1   0 
7 8 7 4 9
the following matrices(when defined)
(a) A · b
(b) a + 4
(c) b · a
(d) a · bT
(e) A · aT
(2) Explain any difference between the answers that MATLAB returns when you type
in A*A, A^2 and A.^2
(3) What is the command that isolates the submatrix that consists of the 2nd to 3rd
rows of the matrix A?
(4) Consider
 
0 4 8 12 16 20 24 28
A =  69 68 67 66 65 64 63 62 
 

1.4 1.1 0.8 0.5 0.2 −0.1 −0.4 −0.7

What is the Matlab command that creates a 3x4 submatrix B from the 1st, 3rd, and
4th rows, and the 1st, 3rd, 5th, and 7th columns of the matrix A.
(5) The depth of a well in meters, d, can be calculated from the time it takes for a stone
that is dropped into the well (with zero initial velocity) to hit the bottom by the
equation
1 2
d = gt
2
where t is time in seconds and g = 9.81 (m/sec2 ), which is the gravitational acceler-
ation . Determine d for the impact times t = 1, 2, 3, ..., 10
(6) For x = [2:2:10] and y = [3:3:15] , calculate z using element-by-element calculation
y
xy+ x
if z = (x+y)(y−x) + 12x
y
 2.10. PROBLEMS 23

(7) Estimate numerically that limn→∞ 1 + n 1 n

= e. First create a vector n as n = [1

10 100 500 1000 2000 4000 8000]. Then, create a new vector y in which each entry
of the vector is calculated by y = 1 + n1 n from each n. Compare the elements

of y with the known value of e ≈ 2.71828. Show numerically that as nincreases, y

approaches the limit of exp(1), or e.
 CHAPTER 3

FUNCTIONS and SCRIPTS in MATLAB

3.1. Inline functions

An easy way of defining functions in MATLAB is inline functions:

Functions with one variable.

>‌>f=@(x) x^2+1
f =
@(x)x^2+1
>‌> f(2)
ans =
5

Functions with two variables.

>‌>f=@(x,y) x^2+y/5
f =
@(x,y)x^2+y/5
>‌>f(1,5)
ans =
2

Function with three variables.

>‌>f=@(x,y,z) 1/z+x^2+y/5;
>‌>f(1,2,5)
ans =
1.6000
24
 3.3. PLOT FUNCTIONS 25

Function with two variables (alternative).

>‌>f=@(x) x(1)*x(2)+x(1);
>‌>f([3 4])
ans =
15
>‌>f=@(x) x(1)+x(2)/7+x(3)^4;
>‌>f([1 1 1])
ans =
2.1429

3.2. Saving and Loading your work in an “.m” file

• If you type
>‌>save myfile
in the command window, all defined variables will be saved to a file called myfile.mat
in the current directory. This file format is specific to MATLAB. You can also select
a subset of variables to be saved by typing their names after the file name such as
>‌>save myfile a,b,C
• If you later enter load myfile, the saved variables are returned to the workspace,
overwriting any presently defined values assigned to the same names.

3.3. Plot functions

ezplot : Easy to use 2D-plot

plot : 2D plotter
plot3d : 3D plotter
image : Displays image of a matrix
surf : 3-D colored surface.
surfc : 3-D colored surface.

Plotting with ezplot. ezplot makes two-dimensional (2D) plots of explicit, implicit
or parametric functions.

• Explicit functions: y = f (x)

• Implicit functions: F (x,
(y) = 0
x = f (t)
• Parametric functions:
y = g(t)
 3.3. PLOT FUNCTIONS 26

ezplot for explicit functions. ezplot(fun) plots the function fun(x) over the default
domain −2π < x < 2π, where y=fun(x) is an explicitly defined function of x.

>‌> ezplot(@sin)

sin

0.5

−0.5

−1

−6 −4 −2 0 2 4 6
x

ezplot for explicit functions: defining a custom domain.

ezplot(@sin,[0 10*pi])

sin

0.5

−0.5

−1

0 5 10 15 20 25 30
x

ezplot for implicit functions. ezplot(fun) plots the implicitly defined function fun(x,y)
= 0 over the default domain −2π < x < 2π and −2π < y < 2π.

>‌> ezplot(@(x,y) x.ˆ4+y.ˆ4-225)

x4+y4−225 = 0
6

0
y

−2

−4

−6
−6 −4 −2 0 2 4 6
x
 3.4. PLOT MULTIPLE CURVES IN 2D WITH PLOT: VECTOR VS MATRIX 27

ezplot for parametric function.

ezplot(fx,fy,[tmin ,tmax]) plots f x(t) and f y(t) over tmin < t <
tmax .
>‌> x= @(t) exp(-t).*cos(8*t);
>‌> y= @(t) exp(-t).*sin(8*t);
>‌> ezplot(x,y,[0 3])
x = exp(−t) cos(8 t), y = exp(−t) sin(8 t)
0.8

0.6

0.4

0.2
y

−0.2

−0.4

−0.6 −0.4 −0.2 0 0.2 0.4 0.6 0.8 1

x

3.3.1. Plot curves in 2D with “plot” function. plot(X,Y) plots vector Y versus
vector X. If X or Y is a matrix, then the vector is plotted versus the rows or columns of the
matrix, whichever line up.

>‌> x = -2*pi:(pi/10):2*pi;
>‌> y1 =sin(x);
>‌> y2 =cos(x);
>‌> plot(x,y1,x,y2)
1

0.8

0.6

0.4

0.2

−0.2

−0.4

−0.6

−0.8

−1
−8 −6 −4 −2 0 2 4 6 8

3.4. Plot multiple curves in 2D with plot: Vector vs Matrix

Example 8. X and Y are both matrices, then they must have equal size. plot(X,Y) plots
columns of Y versus columns of X. If Xis a vector , then the plot function plots each matrix
column Y versus X.

>‌> x=[0:.1:1];
>‌> A=[1:11;10:-1:0]
 3.5. PLOTTING CURVES IN 3D WITH PLOT3 28

>‌> plot(x,A)

12

10

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

3.5. Plotting curves in 3D with plot3

• plot3(x,y,z), where x, y and z are three vectors of the same length, plots a line in
3-space through the points whose coordinates are the elements of x, y and z.

>‌> t = 0:pi/50:10*pi;
>‌> plot3(sin(t),cos(t),t); %parametric functions

40

30

20
t

10

0
1
0.5 1
0 0.5
0
−0.5 −0.5
−1 −1
cos(t) sin(t)

>‌> x=[0:.1:5];
>‌> y=2*[0:.1:5];
>‌> f=@(x,y) sin(x).^2+cos(y).^2; % explicit functions
>‌> plot3(x,y,f(x,y))

1.5
f(x,y)

0.5

0
10
5
4
5 3
2
1
0 0
y x
 3.7. PLOTTING IN 3D WITH SURF 29

3.6. Plotting in 3D with plot3

plot3(X,Y,Z), where X1, Y1, Z1 are vectors or matrices, plots one or more lines in three-
dimensional space through the points whose coordinates are the elements of X, Y, and Z.

Example 9. Creating a meshgrid

>‌> x1=0:4;
>‌> x2=0:5;
>‌> [x,y]=meshgrid(x1,x2) % creating a meshgrid

Example 10. Plot z = sin(xy) cos(y) when −1 ≤ x ≤ 1 and −3 ≤ y ≤ 2

>‌> [x,y]=meshgrid(-1:.1:1,-3:.02:2);
>‌> z=sin(x.*y).*cos(y);
>‌> plot3(x,y,z)

0.5
z=f(x,y)

−0.5

−1
2
1
0
0.5
−2 0
−0.5
−4 −1
y x

3.7. Plotting in 3D with surf

surf(X,Y,Z) plots the colored parametric surface.

>‌> [x,y]=meshgrid(-1:.1:1,-3:.02:2);
>‌> z=sin(x.*y).*cos(y);
>‌> surf(x,y,z)
 3.9. IMAGE 30

0.5

−0.5

−1
2
1 1
0 0.5
−1 0
−2 −0.5
−3 −1

3.8. Plotting in 3D with surfc

surfc: : Combination surf/contour plot.

>‌> [x,y]=meshgrid(-1:.1:1,-3:.02:2);
>‌> z=sin(x.*y).*cos(y);
>‌> surfc(x,y,z)

0.5

−0.5

−1
2
1 1
0 0.5
−1 0
−2 −0.5
−3 −1

3.9. image

The image command is useful for large matrices which would take too much computing time
to plot

image(C): : displays matrix C as an image.

imagesc(C): : Scale data and display as image.

>‌> A=[1,2;3,4];
>‌> imagesc(A),colorbar
 3.10. PROBLEMS 31
0.5 4

3.5

1.5 2.5

1.5

2.5 1
0.5 1 1.5 2 2.5

3.10. Problems
1
(1) Compute 2 + 3, 2 × 3,2/3, 23 , 2 3 , log(3), cos(3), sin(5) in the command line.
(2) Plot f (x) = e−x on [−2, 2].
(3) Plot g(t) = |t| on [−1, 1].
(4) Use linspace to generate a row vector of 100 linearly evenly spaced points between
-1 and 1 and evaluate f (x) = x2 at each point. Plot the graph.
(5) Make surface plots of the following functions over the given ranges:
(a) (x2 + 3y 2 )e−x −y , −3 ≤ x ≤ 3, −3 ≤ y ≤ 3.
2 2

(b) −3y/(x2 + y 2 + 1), |x| ≤ 2, |y| ≤ 4.

(c) |x| + |y|, |x| ≤ 1, |y| ≤ 1.
 CHAPTER 4

WRITING YOUR OWN FUNCTIONS and SCRIPTS in MATLAB

An M-file is a plain text file containing MATLAB commands and saved with the file name
extension .m. There are two types:

• Scripts
• Functions

4.1. Driver codes and functions in MATLAB

Driver Code (MATLAB Script): sends inputs to the function and gets back output

Function Code (MATLAB function): takes in inputs from driver and runs an
algorithm and sends back outputs to the driver

32
 4.1. DRIVER CODES AND FUNCTIONS IN MATLAB 33

Reminder : an extremely important type of statement in any M-file is the comment, which
is indicated by a percent sign %. Any text on the same line after a percent sign is ignored.
Furthermore, the first block of comments in an M-file serves as documentation for the file
and will be typed out in the command window if the help command is used on the file.

Structure of a MATLAB Script.

% Some comments go here

% .....
% .....
Input

Body of Algorithm

Output

Example 11. Open a new MATLAB editor page and type the following MATLAB commands

% This MATLAB scripts multiplies a 3x3 random matrix

% by a 3x1 random vector
clc
clear all
A=rand(3)
b=rand(1,3)
c=A*b’

and save it under a name Driver.m. Then go to the command window type Driver and push
ENTER. What happens?

Structure of a MATLAB Function.

function Output=FunctionName(Input)

Body of Algorithm

Output=...;

Typical MATLAB function with starts with a line such as

 4.2. if...elseif...else...end CONDITIONAL STATEMENTS 34

function [output1,output2] = myfun(input1,input2,input3)

Example 12. Take the script in Example 11 and convert it into a MATLAB function. Your
function will have two inputs: (1) 3x3 random matrix, (2) 3x1 random vector. It will have
one output, c, which is defined as c = A · bT
function c=fcn(A,b)
% This MATLAB scripts multiplies a 3x3 random matrix
% by a 3x1 random vector
c=A*b’;
Example 13. Write a set of a function and a driver that implements the quadratic formula
for finding the roots of ax2 + bx + c = 0

ax2 + bx + c = 0
−b ± d
x1,2 =
2a
√
d = b2 − 4ac
% Driver
a=1;
b=4;
c=3;
[x1,x2] = quadform(a,b,c)

In a separate file:

function [x1,x2] = quadform(a,b,c)

d = sqrt(bˆ2 - 4*a*c);
x1 = (-b + d)/(2*a);
x2 = (-b - d)/(2*a);
Example 14. Write a MATLAB function that computes base 3 logarithm of any given
number by using the change-of-base formula
function a=log3(x)
% This code computes the base 3 logarithm of x
a=log(x)/log(3);

4.2. if...elseif...else...end conditional statements

These statements are executed or not depending on whether a relation holds true.
 MATLAB’S RELATIONAL AND LOGICAL OPERATORS 35

Structure of “if elseif...elseif...else...end” statement.

if (relation)
statement(s)
elseif (relation)
statement(s)
elseif (relation)
statement(s)
.
.
.
else
statement(s)
end

The simplest if...end statement looks like

if (relation)
statement(s)
end

MATLAB’s Relational and Logical Operators

• < : less than

• > : greater than
• <= : less than or equal to
• >= : greater than or equal to
• == : equal to
• ~= : not equal to

Example 15. An example

if (5>3)
disp(’5 is bigger than 3’)
end

Example 16. MATLAB function with an if..else..if statement that computes log3 (x) for
a given number x > 0
 MATLAB’S RELATIONAL AND LOGICAL OPERATORS 36

function a=log3(x)
% This code computes the base 3 logarithm of x
a = NaN;
if sum(-x>0)==0
a=log(x)/log(3);
else
disp(’x has negative value(s)’)
end

Example 17. The Newton-Raphson method (also known as Newton’s Method) for finding
roots of the equation f (x) = 0 is

f (xn )
xn+1 = xn −
f 0 (xn )
To run this method, an initial guess x0 has to be provided. Now, write a MATLAB function
√
that implements this method to approximate 5 in 4 steps

If f (x) = x2 − 5 ⇒ f 0 (x) = 2x. Hence the formula becomes

xn 5
xn+1 = +
2 2xn
Given x0 , we want to compute x4 .

% Driver part (Filename: Driver.m)

x4=NewtonAlg(x0)

% function part(Filename: NewtonAlg.m)

function [x4]=NewtonAlg(x0)
x1=x0/2+5/(2*x0);
x2=x1/2+5/(2*x1);
x3=x2/2+5/(2*x2);
x4=x3/2+5/(2*x3);

Example 18. MATLAB function for f (x) = |x|

 4.4. STRUCTURE OF for...end LOOP 37

function y = AbsVal(x)
y=x;
if x<0
y=-x;
end

4.3. Loops in MATLAB

There are two basic kinds of loops

(1) for...end loops

(2) while...end loops

4.4. Structure of for...end loop

Statements (indexed by j) are executed step by step first through last.

for j=first:last
(statements)
end

Example 19. The following script prints the first four positive integers

for i=1:4
i
end

Example 20. The following script squares each entry in a given vector v with length 10

v=1:10 % define v here

for i=1:10
v2(i) = v(i)^2;
end

Example 21. The Newton-Raphson method for the same function f (x) = x2 − 5 with a
for..end loop in 4-steps is given below
 4.4. STRUCTURE OF for...end LOOP 38

% Driver part
x0=1;
RootOut=NewtonAlg(x0)

% function part
function [x_out]=NewtonAlg(x_in)
for i=1:4
x_out = x_in/2+5/(2*x_in) ;
x_in = x_out;
end

4.4.1. Nested for..end loops.

for j=j0:jn
(statements)
for k=k0:kn
(statements)
end
end

Example 22. The code

for i=1:2
for j=1:3
[i,j]
end
end

will produce the following output

i = 1 1 1 2 2 2
j = 1 2 3 1 2 3

Example 23. A function to compute the transpose of a given nxn square matrix A

function B=transposeA(A)
%This code computes transpose of a matrix
[m,n] = size(A);
for i=1:n
for j=1:n
B(i,j)=A(j,i);
 4.4. STRUCTURE OF for...end LOOP 39

end
end

Example 24. Write a function that adds up all the entries in a given mxn matrix A

function [s_sum]=Sum_of_A(A)
% This function computes sum of all entries in a matrix
[m,n] = size(A);
s_sum=0
for i=1:m
for j=1:n
s_sum=s_sum+A(i,j);
end
end

Example 25. A function that produces an upper triangular mxn matrix B from an mxn
matrix A. In other words,

   
a11 a12 a13 ... a1n a11 a12 a13 ... a1n
a21 a22 a23 ... a2n 0 a22 a23 ... a2n
   
   
   
A= a31 a32 a33 ... a3n ⇒B= 0 0 a33 ... a3n 
.. .. .. .. .. .. .. .. .. ..
   
. . . . . . . . . .
   
   
am1 am2 am3 ... amn 0 0 0 ... amn

Answer:

function [B]=Upper(A)
[m,n]=size(A);
B=zeros(m,n);
for i=1:m
for j=i:n
B(i,j)=A(i,j);
end
end
h i
Example 26. The dot product of two vectors with same sizes v = v1 v2 ... vn and
h i
w = w1 w2 ... wn are defined as
 4.5. STRUCTURE OF while...end LOOP 40

n
X
vw = vi wi = v1 w1 + v2 w2 + ... + vn wn
i=1

The l2 −norm of v is defined as

v
u n q
uX
|v| = t vi = v12 + v22 + ... + vn2
2

i=1

Exercise. Write a function that takes in v and w and sends out their dot product and
l2 −norms.

4.5. Structure of while...end loop

Statement is executed while the boolean expression remains true.

while (expression)
(statement)
end

Example 27. The following code prints the first four positive integers and employs a
while...end loop

j=1;
while j<=4
j
j=j+1;
end

Example 28. The following code computes roots of x2 − 5 = 0 , until numerical error
becomes less than given a preset error toleranceEr and uses while...end

% Driver part
clc
Er = 1e-8;
x_out = 0;
x_in = 1;
[x_out] = NewtonAlg(x_in,Er)
% function part
function [x_out]=NewtonAlg(x_in,Er)
while abs(x_in-sqrt(5))>Er
 4.5. STRUCTURE OF while...end LOOP 41

x_out = x_in/2+5/(2*x_in) ;
x_in = x_out;
end

Example 29. Determine what the following MATLAB function does. Then explain the
output of [a] = exple(10)

function [a] = exple(n)

a = 0;
while 2^a < n
a = a + 1;
end % End of function

Example 30. The code given below computes S =

P100
n=1 n

% Sum of first 100 positive integers

tsum=0;
for i=1:100
tsum=tsum+i;
end
tsum

Example 31. Generate 2000 random numbers between 0 and 2000 and count how many of
them are in the following intervals. Then use MATLAB’s bar built-in function to plot the
counts you computed to produce a histogram.

• (a) [0.00,0.25)
• (b) [0.25,0.50)
• (c) [0.50,0.75)
• (d) [0.75,1.00]

Answer:

A = rand(1,2000);
count = zeros(1,4);
for i = 1:2000
if (A(i)>=0.0 && A(i)<0.25)
count(1)=count(1)+1;
elseif (A(i)>=0.25 && A(i)<0.50)
 4.5. STRUCTURE OF while...end LOOP 42

count(2)=count(2)+1;
elseif (A(i)>=0.50 && A(i)<0.75)
count(3)=count(3)+1;
else
count(4)=count(4)+1;
end
end
√
Example 32. Write a function that approximates 5 using Newton’s Method in n steps in
two different ways:

(a) without using an array

√
(b) using an array to store all approximations to 5

Newton Algorithm 2 (using for loop)

function [x_out] = NewtonAlg2(x_in,n)

for i = 1:n
x_out = x_in/2 + 5/(2*x_in);
x_in = x_out;
end

Newton Algorithm 3 (using array)

function [xfinal] = NewtonAlg3(x_in,n)

x=zeros(1,numIter+1);
x(1) = x_in;
for i = 1:n
x(i+1) = x(i)/2 + 5/(2*x(i));
end
xfinal = x(i+1);

Example 33. Write a MATLAB script with a for..end loop and if..else..end statements
that plots the graph of the following piecewise function


x 2 if − 1 ≤ x < 0.5
f (x) =
0.25 if 0.5 ≤ x ≤ 1

Answer:....
 4.5. STRUCTURE OF while...end LOOP 43

x=[-1:.1:1];
for i=1:length(x)
if x(i)<0.5
f(i)=x(i)^2;
else
f(i)=0.25;
end
end
plot(x,f)
xlabel(’x’)
ylabel(’f’)

Example 34. Develop a MATLAB function that solves the linear system Ax = b for a given
nxn square matrix A and a nx1 vector b (Use if...end statements to check if the system has
a solution first)
 4.6. PROBLEMS 44

4.6. Problems

Answer the following questions

(1) Write a MATLAB function that computes n! = n × (n − 1) × ... × 3 × 2 × 1 (Use

if...end and for...end statements)

(2) Write a MATLAB function that computes k=1 k2 .

Pn 1

(3) The Fibonacci sequence Fn = Fn−1 + Fn−2 for n ≥ 2, where F0 = 0; F1 = 1. Write

a MATLAB function that computes the nth Fibonacci number.

(4) Develop a driver called Driver_Quad.m and two other MatLab functions called
R_Quad.m and Disc_Quad.m that solves a quadratic equation

ax2 + bx + c = 0

Organize your codes so that the driver code Driver_Quad.m supplies the coefficients
a, b and c into the function file R_Quad.m, that computes the roots x1 and x2 from
the quadratic formula and sends them back to the driver Driver_Quad.m. While
computing the roots in R_Quad.m, have it call another function file Disc_Quad.m
to get the discriminant value calculated.
(5) Plot the graphs of f (x) = x2 , g(x) = x3 for x = −1, . . . , 1 on the same axis. Label
the x and y axes and create a legend indicating which graph is which.
(6) Let x = [ 3, 2, −1, 5, 8 ], determine outputs of each of the following MATLAB
command
(a) x(end)
(b) x(3)
(c) length(x)
(d) x(2:4)
(7) A magic square is an n×n matrix in which each integer 1, 2, . . . , n2 appears once
and for which all the row, column, and diagonal sums are identical. MatLab has a
command magic that returns magic squares. Check its output when n=5 and n=20
and use MatLab to verify the summation property. (The antidiagonal sum will be
the trickiest. Look for help on “fliplr” .)
(8) Let P1, P2, and P3 be the vertices of an equilateral triangle. Start with a point
anywhere inside the triangle. At random, pick one of the three vertices and move
halfway toward it. Repeat indefinitely. If you plot all the points obtained, a very
clear pattern will emerge. (Hint: This is particularly easy to do if you use complex
numbers. If z is complex, then plot(z) is equivalent to plot(real(z),imag(z)).)
 4.6. PROBLEMS 45

(9) Generate 100 random matrices using randn(100), and plot all of their eigenvalues
as dots in the complex plane on one graph. (Thus, you should see 10,000 dots.)
Use axis equal to make the aspect ratio one-to-one. You should see a fairly striking
result.
 CHAPTER 5

SOME BUILT-IN FUNCTIONS in MATLAB

5.1. roots: Polynomial roots

roots computes the roots of the polynomial

whose coefficients are the elements of
the vector C. If C has N +
1 components, then the polynomial is
C1 X N + ... + CN X + CN +1

Example 35. Solve for x

x2 + 3x + 2 = 0,

roots([1 3 2])
ans=
-2
-1

Example 36. Solve for x

x5 + x4 + x3 + x2 + x + 1 = 0,

>‌> S=roots([1 1 1 1 1 1 ])
S =
0.5000 + 0.8660i
0.5000 - 0.8660i
-1.0000
-0.5000 + 0.8660i
-0.5000 - 0.8660i

Example 37. Lets solve

x2 + 3x + 5 = 0,

>‌>S= roots([1 3 5])

S =
46
 5.2. fsolve: SOLVING SYSTEM OF NONLINEAR EQUATIONS WITH SEVERAL VARIABLES 47

-1.5000 + 1.6583i
-1.5000 - 1.6583i
>‌> real(S)
ans =
-1.5000 -1.5000
>‌> imag(S)
ans =
1.6583 -1.6583

Example 38. Look at another example

2x5 − 13x2 + 5x + 1 = 0

>‌>roots([2 0 0 -13 5 1])

ans =
-1.0437 + 1.6365i
-1.0437 - 1.6365i
1.6925
0.5401
-0.1452

5.2. fsolve: Solving system of nonlinear equations with several variables

fsolve attempts to solve equations like

F(X) = 0
where F and X may be vectors or matrices.

Example 39. Solve x − ex = 0, x0 = 2

>‌> f=@(x) x-exp(x);

>‌> fsolve(f,2)
ans =
3.0136e-07
% Alternative way:
function Driver
x =fsolve(@fcn,2)
function f = fcn(x)
f = x-exp(x);
 5.3. fminbnd SINGLE-VARIABLE BOUNDED NONLINEAR FUNCTION MINIMIZATION 48

Example 40. Solve the following system numerically using fsolve

(
x2 − xy + y 2 2 = 1
3xy − y 3 = 2

Solve the following system numerically using fsolve

(
2x − y = e−x
−x + 2y = e−y

5.3. fminbnd Single-variable bounded nonlinear function minimization

X = fminbnd(FUN,x1,x2) attempts to find

a local minimizer X of the function FUN
in the interval x1 < X < x2.

Example 41. Mimimize the following function over −5 ≤ x ≤ 5

f (x) = sin x

clear all
clc
clf
%% List of functions
fcn=@(x) sin(x);
[xout fval] =fminbnd(fcn,-5*0,1)
plot(xout,fval,’ro’,’markersize’,20),
hold on
u=[-3:.01:3];
plot(u,fcn(u))

Exercise 42. Use the code above to minimize the following functions

• f (x) = sin x − 0.6

• f (x) = x2
• f (x) = (x−0.3)12 +0.01 + 1
((x−0.9)2 +.04)−6)
 5.4. fminsearch MULTIDIMENSIONAL UNCONSTRAINED NONLINEAR MINIMIZATION 49

5.4. fminsearch Multidimensional unconstrained nonlinear minimization

X = fminsearch(FUN,X0) starts at X0 and

attempts to find a local minimizer X of
the function FUN.

Example 43. Minimize the following function (called the Rosenbrock Valley Function) start-
ing from x0 = (2, 3)

f (x1 , x2 ) = 100(x2 − x1 )2 + (1 − x1 )

%% Rosenbrock Valley
f =@(x) 100*(x(2)-x(1)^2).^2+(1-x(1)).^2;
[xout fval] =fminsearch(f,[2 2])
%% Plotting Rosenbrock Valley
[X,Y]=meshgrid(-3:0.2:3,-3:0.2:3);
surfc(X,Y,f(X,Y))

Example 44. Minimize the following function (Rasting Function) starting from (2, 3)

f (x1 , x2 ) = 20 + 10 cos(2πx1 ) + 10 cos(2πx2 )

%% Rastrigin function
[xout fval] =fminsearch(f,[2 2])
f =@(x) 20+10*cos(2*pi*x(1))+10*cos(2*pi*x(2));
%% Plotting Rastrigin function
[X,Y]=meshgrid(-2:0.02:2,-2:0.02:2);
Z = 20 +10*cos(2*pi*X)++10*cos(2*pi*Y);
surfc(X,Y,Z)

Example 45. (Data fitting)Codes that mimimize the sum of square differences between the
data and a given function

%% ----------- Driver function ---------------

function fminsearchApplication
%% Generating artificial data
clc
a = 1;
5.4. fminsearch MULTIDIMENSIONAL UNCONSTRAINED NONLINEAR MINIMIZATION 50

b = 1;
c = 1;
xdata = -5:.5:5;
per = 30/100;
ydata = fcn(xdata,a,b,c).*(1+per*randn(1,length(xdata)));
%% Estimate a,b,c
a0 = rand;
b0 = rand;
c0 = rand;
x0 = [a0 b0 c0];
[xout fval] = fminsearch(@sum_square,x0,[],xdata,ydata)
%% ------------ Objective function -------
function SSE= sum_square(p,xdata,ydata)
a = p(1);
b = p(2);
c = p(3);
%% Definition of the objective function
M = fcn(xdata,a,b,c);
SSE = sum((ydata-M).^2);
% Plot every step plot(xdata,ydata,’ro’,xdata,M,’b-s’),
axis([2*min(xdata) 2*max(xdata) 4*min(ydata)-2 2*max(ydata)])
pause(.01)
%% ------------ Model --------------------
function fout = fcn(xdata,a,b,c)
fout = a*sin(b*xdata)+c;
 5.6. MATLAB: USING THE DEBUGGER 51

5.5. lsqcurvefit: solves non-linear least squares problems

X = lsqcurvefit(FUN,X0,XDATA,YDATA)
starts at X0 and finds
coefficients X to best fit
the nonlinear functions

Example 46. Following code minimize the difference between data and the function

f = a cos(bx) + c

% Driver
clc
% Input data set
xdata=[-5.0 -4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0 5.0];
ydata =[8.5 10.8 11.9 11.5 9.7 7.0 4.3 2.4 2.0 3.2 5.5];
% Initial estimate
X0=2+rand(1,3);
% Upper and lower bounds
LB=[];
UB=[];
% call lsqcurvefit
[xout,resnorm] = lsqcurvefit(@fcn,X0,xdata,ydata,LB,UB,[],ydata)
% the optimization function
function f=fcn(X0,xdata,ydata)
a=X0(1);
b=X0(2);
c=X0(3);
f=a*cos(b*xdata)+c;
% plotting for every estimnate
plot(xdata,ydata,’ro’,xdata,f,’b’)
pause(.01)

5.6. MATLAB: Using the Debugger

The debugger helps us understand what a program actually does (rather than what we want
it to do)
 5.7. MATLAB HINTS (SOURCE: TOBIN A. DRISCOLL 2009) 52

• Setup breakpoints in an m-file by pressing F12, clicking on the file icon with the red
dot, or clicking on the "-" in the margin of the editor.
• Execute the m-file at the command line and note the prompt change to K>‌>.
• dbstep executes the next command in the program (after the breakpoint or previous
command)
• dbcont executes all remaining steps of the program until the next breakpoint or
termination of the program.
• dbquit exits debugging mode

5.7. MATLAB Hints (Source: Tobin A. Driscoll 2009)

A few things things about MATLAB that can come in handy, but which often do not come
to the attention of beginners:

• Use the up-arrow key to cycle through previous commands. If you type specific
characters first, only commands matching the typed characters will be recalled.
• If a computation is taking too long, interrupt it by pressing Ctrl-C (after making
sure the Command Window is active in the operating system).
• Even when MATLAB displays only 4–5 digits of a result, it’s storing about 15
significant digits. (You can see them all by typing format long). By copying or
retyping a displayed result, you throw away a lot of information. Wait until the end
of the calculation to round off your results.
• MATLAB has great debugging tools. Run your code step by step to uncover errors.
Run someone else’s code step by step to understand it thoroughly.
• The code checker (checkcode) makes some good suggestions
• The previous two items alone are sufficient reasons to use the built-in MATLAB
Editor for writing code. Open it by entering edit.
• If the execution of your code is too slow to suit, use the Profiler to find the slowest
steps.
• Don’t use a screen or window capture function to paste figures into a document
or presentation. The results look cheesy and amateurish, MATLAB’s use print
command (look at the help file to get more about usage of this command)
• After you have properly exported a figure to a graphics file, save that figure again
in the native FIG format. You may want to make changes to it someday.
 5.8. PROBLEMS 53

5.8. Problems

Explain what each of the following MATLAB codes does when it is run. Write next to each
line what that line does

%% ===================================================’)
%% Code 1
%% ===================================================’)
clear all
clc
clf
A = rand(1,1000);
C = zeros(1,5);
for I = 1:length(A)
if A(i)<=.2
C(1) = C(1)+1;
end
if A(i)>.2 & A(i)<=.4
C(2) = C(2)+1;
end
if A(i)>.4 & A(i)<=.6
C(3) = C(3)+1;
end
if A(i)>.6 & A(i)<=.8
C(4) = C(4)+1;
end
if A(i)>.8 & A(i)<=1
C(5) = C(5)+1;
end
end
xx = [1:5];
plot(xx,C)
%% ===================================================’)
%% Code 2
%% ===================================================’)
clear all
clc
clf
 5.8. PROBLEMS 54

A = rand(1,100);
L = 1;
for j=1:length(A)/10
K(L,1:10) = A(10*(j-1)+1:10*j);
L = L+1;
end

%% ===================================================’)
%% Code 3
%% ===================================================’)
clear all
clc
clf
A = pi*100*ones(1,5);
sprintf(’ %f \n %.2f \n %+.2f \n %12.2f \n %012.2f \n’, A)
sprintf(’|%e|%15e|%f|%15f|’, pi*50*ones(1,4))
%% ===================================================’)
%% Code 4
%% ===================================================’)
clear all
clc
clf
R = 2.9;
t = [0:.1:1];
Rx = [R*t.*(1-t)];
subplot(2,1,1),plot(t,t,’b’), hold on;
subplot(2,1,1),plot(t,Rx ,’r’), hold on;
N = length(t);
x(1) = 0.6;
for i=1:length(t)
x(i+1) = R*x(i)*(1-x(i));%
end
plot(t,x)
%% ===================================================’)
%% Code 5
%% ===================================================’)
clear all
 5.8. PROBLEMS 55

clf
clc
%% ====================== Part(A) ===================
R = 2.9;
t = [0:.1:1];
N = length(t);
x(1)= 0.6;
for i=1:length(t)
x(i+1) = R*x(i)*(1-x(i));%
end
subplot(2,1,1),plot(t,t,’b’), hold on;
subplot(2,1,1),plot(t,R*t.*(1-t),’r’), hold on;
axis(’square’);
axis([0 1 0 1]);
set(gca,’XTick’,(0:.1:1),’YTick’,(0:.1:1))
grid on;
xlabel(’xt’)
ylabel(’xt+1’)
%% ====================== Part(B) ===================
subplot(2,1,1),line([x(1) x(1)],[0 x(2)],’Color’,’g’)
subplot(2,1,1),plot(x(1), x(1),’ko’);
for ic=1:length(t)-2
subplot(2,1,1), line([x(ic) x(ic+1)],[x(ic+1) x(ic+1)],’Color’,’g’)
subplot(2,1,1), line([x(ic+1) x(ic+1)],[x(ic+1) x(ic+2)],’Color’,’g’)
subplot(2,1,1), plot(x(ic+1), x(ic+1),’ko’);
pause
end
line([x(N) x(N+1)],[x(N+1) x(N+1)],’Color’,’g’)

%% ====================== Part(C) ===================

At = text(.1,.6,[’R=’,num2str(R)]); set(at,’FontSize’,12);
pause(20)
subplot(2,1,2),plot(t,x(1:end-1),’b-o’)
axis(’square’);
axis([0 1 0 1]);
xlabel(’t’)
ylabel(’xt’)
 CHAPTER 6

SOLVING ORDINARY DIFFERENTIAL EQUATIONS in

MATLAB

A MATLAB code for solving a differential equation with an initial condition (IVP) usually
consists of two parts: (1) driver part and (2) differential equation part. The driver part
constitutes initial values for dependent variables, parameter values, one of MATLAB’s dif-
ferential equation solver and a common for plotting the solution, whereas the function part
consists of the differential equation. The general form of an initial value problem (IVP) is

dx
= f (t, x)
dt
x(t0 ) = x0

To solve this problem with the right hand side of the differential equation f = f (t, x), the
initial starting point (t0 , x0 ) and a range [t0 , tend ] for the independent variable t have to be
provided. There are a number of ODE solvers in MATLAB, each using a different numerical
method. Here, our focus is on the solver ode15s.

6.1. Solving an ODE with an inline function

This is the simplest way of solving an ODE with an initial condition is to use inline function.
Consider the following example

Example 47. Numerically approximate the solution of the first order differential equation
given below

dx
= 0.5x, and x(0) = 1.0
dt
on the interval x ∈ [0, 5]. We begin by defining the function f (t, x) in dx
dt
= f (t, x):

fcn=@(t,x) -.5*x;

The basic usage for MATLAB’s solver ode15s is

ode15s(function,domain,initial condition).
That is, we use
56
 6.2. SOLVING AN ODE WITH “.M” SCRIPT AND A FUNCTION 57

[Tout,Xout]=ode15s(f,[0 .5],1);
plot(Tout,Xout)
xlabel(’t’)
ylabel(’x(t)’)

and MATLAB returns two column vectors, the first with values of Tout and the second with
values of Xout(solution). Since the MATLAB output is long and it is omitted here. Now we
can plot the results

plot(Tout,Xout)
xlabel(’t’)
ylabel(’x(t)’)

6.2. Solving an ODE with “.m” script and a function

Example 48. Solve the following IVP for 0 ≤ t ≤ 10

dx
= k1 ln(x) + x − 1
dt
x(0) = 0.8

if k1 = −0.4.
MATLAB code for this problem is given below.

%% ------------------------ Driver part -----------------

clc % Clears Command Window
clf % Clears figure Window
%% Enter parameters
k1= -.4;
%% Input initial and final time points (Tint and Tend )
T_int = 0;
T_end = 10;
%% Enter Initial Values for the dependent variables
x1_0 = 0.8;
%% Call MATLAB’s solver
[Tout,Xout] = ode15s(@fcn,[T_int T_end],[x1_0],[],k1);
%% Plot the solution
plot(Tout,Xout,’r’)
 6.3. SOLVING A SYSTEM ODES 58

%% Specify labels for x and y axis

xlabel(’t’)
ylabel(’x(t)’)

%%------------------ Differential Equation part ------------

function dxout=fcn(t,x,k1)
% Assign variables
x1 = x(1) % assigns x(1)to the variable x1
%% Enter the equation
dx1 = k1*log(x1)+x1-1;
dxout = [dx1]’;

The result is shown in the plot below

Figure 6.2.1. The solution of the problem in Example 48

6.3. Solving a System ODEs

Below is a template you can modify to develop your code to solve a system of differential
equation involving a number of parameters:
 6.3. SOLVING A SYSTEM ODES 59

%% ------------------------ Driver part -----------------

function Driver_fcn
clc % clears Command Window
clf % clears figure Window
%% Enter parameter values for k1,k2,k3,...
k1= ?;
k2= ?;
k3= ?;
.
.
.
%% Input initial and final time points (T_int and T_end)
T_int = ?;
T_end = ?;
%% Enter Initial Values for the dependent variables
x1_0 = ?;
x2_0 = ?;
.
.
.
%% Call MATLAB’s ode15s solver
[Tout,Xout]=ode15s(@fcn,[T_int T_end],[x1_0 x2_0 · · · ],[],k1,k2,k3,· · · );
%% Plot the solution
plot(Tout,Xout,’r’)
%% Specifiy labels for x and y axis
xlabel(’t’)
ylabel(’x(t)’)

%%------------------ Differential Equation part ------------

function dxout=fcn(t,x,k1,k2,k3,...)
%% Assign variables
x1 = x(1) % assigns x(1)to the variable x
x2 = x(2) % assigns x(2)to the variable x
.
.
.
%% Enter the equations
dx1 = f1(x1,x2,...);
dx2 = f2(x1,x2,...);
.
.
.
dxout = [dx1 dx2 ... ]’;
 6.3. SOLVING A SYSTEM ODES 60

Example 49. Solve the following IVP for 0 ≤ t ≤ 10

dx! x1 x2
= ax1 (1 − x1 ) −
dt k + x1
dx2 x1 x2
= b − x2
dt k + x1
x1 (0) = 1 and x2 (0) = 1

The MATLAB script and function solving this system of differential equations are given
below. The figure is shown below the code.

%% ------------------------ Driver part -----------------

clc % Clears Command Window
clf % Clears figure Window
%% Enter parameters
a = 1;
b = 3;
k = 1/2;
%% Input initial and final time points (T_{int} and T_{end} )
T_int = 0;
T_end = 100;
%% Enter Initial Values for the dependent variables
x1_0 = 1;
x2_0 = 1;
%% Call MATLAB’s solver
[Tout,Xout] = ode15s(@fcn,[T_int T_end],[x1_0 x2_0],[],a,b,k);
%% Plot the solution
plot(Tout,Xout)
%% Specifiy labels for x and y axis
xlabel(’time’)
ylabel(’x_1(t) and x_2(t)’)

%%------------------ Differential Equation part ------------

function dout=fcn(t,x,a,b,k)
% Assign variables
x1 = x(1);
x2 = x(2);
 6.3. SOLVING A SYSTEM ODES 61

%% Enter the equation

dx1 = a*x1*(1-x1)-x1*x2/(k+x1);
dx2 = b*x1*x2/(k+x1)-x2;
dout = [dx1 dx2]’;

Figure 6.3.1. The Solution of the problem in Example 49

2.5

2
x1(t) and x2(t)

1.5

0.5

0
0 10 20 30 40 50 60 70 80 90 100
time