Justin’s Guide to MATLAB in MATH 240 - Part 1

This “Justin’s Guide” is an edited version of a guide by Justin Wyss-Gallifent. It is instructional guide
before the actual Project 1, with information you will need to have. You do not need to turn in anything
for this part. Project 1 starts a few pages down.

1. Method
The way this guide is written is that it is assumed that you will sit down at a MATLAB terminal
and start. The commands that you give to MATLAB are given but the output is not. The output is,
however, talked about, with the understanding that you will put in the commands and see the output.
Then you can read about it and keep going.

2. Starting MATLAB
Find a system on campus and run MATLAB. The important window (there will probably be a couple)
will have a bunch of subwindows in it but the major panel on the right has a >> in it. This is the
prompt, it’s where we tell MATLAB what we want it to do.
3. Simple Calculation
For example, to do 4+5 we just type it in and hit “return”
>> 4+5
4. Solving a System
Now let’s do something having to do with MATH 240.
First we’ll solve a system of linear equations. We do this using MATLAB’s solve command. Due
to the quirkiness with which this command operates we have to make sure we interpret the result
correctly. For example, suppose we wish to solve the system

3x1 + x2 − 5x3 = 0
x1 − 5x2 + x3 = 2
x1 + x2 − 5x3 = −1

what we type in is the following.

>> [x1 x2 x3]=solve(’3*x1+x2-5*x3=0’,’x1-5*x2+x3=2’,’x1+x2-5*x3=-1’)
Note the apostrophes (single quotes), necessary because an equation must be treated as a string of
characters and strings of characters in MATLAB are delimited by apostrophes.
Note here that we got three solutions. What would happen if we fed MATLAB a system with not
enough equations? Let’s try!
>> [x1 x2 x3]=solve(’3*x1+x2-5*x3=0’,’x1-5*x2+x3=2’)
If you try this, MATLAB chokes. The reason is that it doesn’t know what to do or how to give the
solution. There are ways for MATLAB to solve this but they demand that we know more than we do,
too. For now we’ll need to deal with the corresponding augmented matrices in order to get MATLAB
to help us solve this system.
5. Putting in a Matrix
We can put in a matrix easily. To put in
 
2 3
5 −1

we do
>> A=[2 3 ; 5 -1]
6. Matrix Operations
Suppose now you put in a matrix and you wish to manually do row operations to it. MATLAB doesn’t
have any functions for doing row operations, instead we have to execute low-level commands to do so.
This is easier than it seems.
Here is an augmented matrix
>> A=[1 -1 -3 7 ; -3 1 4 -16 ; 4 -3 -5 12]
To get this to row echelon form, first we’ll clean out the entries below the upper-left 1. To do this we
need to add 3 times row 1 to row 2. In MATLAB the code is
>> A(2,:)=A(2,:)+3*A(1,:)
Confused? I was! The expression A(rownumber,:) refers to the entire row, so basically this line is
saying:

A(2,:) = A(2,:) + 3 ∗ A(1,:)

| {z } | {z } | {z }
Row2 Row2 3∗Row1

Try it! The result will be given.

Next we’ll add −4 times row 1 to row 3.
>> A(3,:)=A(3,:)+(-4)*A(1,:)
Now you’ll see that we have 0s below the upper-left 1. Next we need to get a 0 in the bottom row
where the 1 is. The easiest way to do this is first to interchange rows 2 and 3.
>> A([2 3],:)=A([3 2],:)
This is a confusing command but it does the job.
Now we add twice row 2 to row 3.
>> A(3,:)=A(3,:)+2*A(2,:)
The result you’ll see is the matrix in row echelon form. Remember this means any rows of 0s are at
the bottom (none in this case) and the leading entries go down and right.
To get to reduced row echelon form, we first have to get 1s for all our leading entries. The third row
is the only problem, it needs to be multiplied by 1/9.
>> A(3,:)=1/9*A(3,:)
This means that row 3 is 1/9 times what it was.
Lastly we must clear out the entries above the leading 1s.
First above the rightmost leading 1.
>> A(1,:)=A(1,:)+3*A(3,:)
>> A(2,:)=A(2,:)+(-7)*A(3,:)
And then above the leading 1 to the left of that one.
>> A(1,:)=A(1,:)+1*A(2,:)
The result is now in reduced row echelon form. Since there are no free variables we can simply read
off the solution: x1 = 3, x2 = 5 and x3 = −3.
7. Skipping to the Reduced Row Echelon Form
Suppose now you want to solve a system of matrices by getting the augmented matrix in reduced row
echelon form but you don’t want to do all that work on the previous page. The rref command does
this in MATLAB. For example if I put in the augmented matrix (corresponding to a system of linear
equations) for the previous page’s problem:
>> A=[1 -1 -3 7 ; -3 1 4 -16 ; 4 -3 -5 12]
and then reduce it
>> rref(A)
we see that MATLAB saves us time compared to all the work we needed to do on the previous page.
Here is another, this one has free variables at the end. We will see that after rref because there are
non-pivot columns corresponding to variables.
>> A=[1 3 4 2;1 -3 0 1]
and then reduce it
>> rref(A)
we get the reduced form. It’s in decimals but clearly it corresponds to the augmented matrix

3
 
1 0 2 | 2
2 1
0 1 3 | 6

Note that the third column is not a pivot column so x3 is free. Hence we have the solution

3
x1 = − 2x3
2
1 2
x2 = − x3
6 3
x3 = free

8. Lastly, note that we can deal with matrices with unknown constants in them but we have to explicitly
tell MATLAB that they are unknown constants. For example, suppose we wish to enter the matrix
 
−2 3 | h
5 −1 | k

First we must tell MATLAB that h and k are to be dealt with symbolically. We do this with the syms
command
>> syms h
>> syms k
And now we can do
>> A=[-2 3 h ; 5 -1 k]
>> rref(A)
MATH 240 – Fall 2018 – MATLAB Project 1 – due in class Tuesday, 9/18.

What to turn in: A paper version of the completed project should be handed in to your TA in class on
the due date (or earlier). Your printout must include all of your commands, as well as the output
produced by your commands. Here are two suggestions on how to do that:
1. You can turn in a printout of the command window of a clean final session. A clean session is one in
which no mistakes were made. Since mistakes inevitably happen, you may want to save the commands
which work correctly (say, copy and paste them into a text editor), and for the final output, enter the
commands in a single session.

2. An alternative, and better, approach is to write a script (.m file) and then “publish” it in MATLAB.
Here, you can enter all the necessary commands in the script, and you can edit them if there is a
mistake. If you do this, be sure to separate your code problem-by-problem. This can be done by either
creating a new .m file for each problem, or by typing %% in the script. This signifies to MATLAB
that it is a new portion of code, and MATLAB separates it accordingly when published. For example,
you can write
%% Problem 1
followed by all your commands for Problem 1 (then the same for Problem 2 and so on).
Regardless of how you present your project, you will need to include items in what you turn in that are
not MATLAB commands. For example, your name. You can do this by typing into the MATLAB prompt
>> your name, preceded by the percent symbol %. MATLAB ignores the “comment” you type on a line
following the symbol %. For example, if your name were Homer Simpson, your first lines could look like

>> % Homer Simpson

>> % MATH 240 Fall 2018 Project 1
>> % Other group members: Ned Flanders, Sideshow Bob

Also, when you begin a new problem part, indicate that on the output with something like

>> % Problem 1(a)

In the project, you will see questions marked with a ?. These require you to type an answer that is not just
a MATLAB command. Type your answer after a %.

Rules for working in groups: You can work in groups of 1, 2, or 3 with anyone in the course who has
the same TA and instructor as you. List all members of your group at the beginning of your project.
Everyone in a group must type up the project individually, and hand it in to the TA. You can’t
learn MATLAB without typing. You should work side-by-side, either in a computer lab or working on your
own laptops. One project from a group will be graded and everyone in the group will get that grade.

But first: Before you begin the project, go through “Justin’s Guide” and type along in MATLAB as you
read. That should be enough to get you through this project. Don’t turn in anything from what you did in
“Justin’s Guide”. To start fresh, you can enter the command clc to clear the command window, and you
can enter clear to clear out all variables that you have stored.

Format: The command format rat will cause your numerical output to appear as fractions. You might find
that convenient or necessary at times. The command format short changes back to the decimal format. The
command format long produces decimal output with more decimal places. Sometimes, certain problems
will insist that you use certain formats.

The project problems begin on the next page.

0. Reread all directions on the previous page that are typed in boldface. Then enter the command clock
into MATLAB.

1. Consider the system of equations



 x1 − 3x2 − x3 = 6
−2x1 + 8x2 + 4x3 = −11

3x1 − 19x2 − 10x3 = 16


(a) Enter it into MATLAB as an (augmented) matrix named A.

(b) Use elementary row operations (as in part 6 of the guide) to reduce it to row echelon form.
(c) Continue using elementary row operations to get to reduced row echelon form.
(d) ? Give the solution of the system.

2. Consider the system of equations

x1 + x2 − x3 − x4 + 5x5 = −8



−6x − 8x + 4x

− 34x5 = −10
1 2 3

 2x1 + 2x2 − 2x3 + x4 + 7x5


 = 11
− 8x3 − 13x4 + 9x5 = −121

4x1

(a) Enter it into MATLAB as an (augmented) matrix named B.

(b) Use elementary row operations (as in part 6 of the guide) to reduce it to row echelon form.
(c) Continue using elementary row operations to get to reduced row echelon form.
(d) Re-enter the original matrix into MATLAB and use the rref command to instantly get the
reduced row echelon form. Make sure it is the same as your previous answer. (From this point
onward, we will make extensive use of the rref command in MATLAB. You do not need to do
row reduction in MATLAB one elementary row operation at a time anymore.)
(e) ? Give the solution of the system in parametric vector form.

3. Start this problem in format short. Enter the matrix

 
3.5 3.5 7.0 6.3
A = 1.8 2.0 1.4 2.0
1.6 6.4 0.8 5.6

(a) Convert it to reduced row echelon form using rref.

(b) Rewrite this matrix with fractions rather than decimals.
(c) ? Write the solution of the system which corresponds to this matrix (as if it were an augmented
matrix). Use fractions, not decimals.

4. Do problem # 3 on p. 87 of the textbook. For each part, be sure to explictly give the appropriate
system of equations (as a comment) before entering the appropriate matrices into MATLAB. Show all
of your necessary MATLAB computations.
5. Suppose you wish to show that
     
4  1 −2 
5 is not in the span of −2 ,  1 
1 3 2
 

(a) Use rref on an appropriate matrix to show this fact.

(b) ? Explain why your calculation in (a) is sufficient.
(c) ? Are the three vectors above linearly independent or linearly dependent? Justify your answer.
     
a 3 −2
6. Suppose you wish to show that is in the span of , for any a and b.
b 4 5

(a) First declare a and b as symbolic variables.

(b) Enter a matrix A which can help you show this and use rref on it.
     
a 3 −2
(c) ? Explicitly give the weights w1 and w2 such that = w1 + w2 . (They should be
b 4 5
given in terms of a and b.)
7. Suppose you wish to show that the set of vectors
         

 1 3 0 1 6 
         
1 1 −1
  ,   ,   ,   , 2
3



 −1   2   2  −3  2 


1 −2 3 −2 1
 

is linearly dependent.
(a) Enter a matrix A which can help you solve this and use rref on it.
(b) ? Explain how you can tell from the rref that the vectors are linearly dependent.
(c) ? Give a nontrivial linear combination of the vectors which yields 0. (Hint: Make a free variable
nonzero.)
(d) ? Find a theorem in §1.7 which allows us to conclude, without doing any of the previous compu-
tations, that this set must be linearly dependent.
(e) ? Does this set of vectors span R4 ? Justify your answer.