Department of Mathematics

Math 300: Mathematical Computing

Matlab Vectors and Matrices

Matlab stands for "Matrix Laboratory", so it is not a surprise that Matlab has simple, intuitive, but powerful matrix structures. The general term for such structures is array, but to retain the link to mathematics we will refer to vectors and matrices. A vector is a collection of values indexed using a single integer, while a matrix is a collection of values indexed by an order pair of integers. We have already seen the most simpleminded way to enter a vector:

v = [1 2 3 4 5 6 7 8 9 10]

However, while simpleminded, this is not the simplest way to enter that vector. The following is a completely equivalent statement in Matlab.

v = 1:10

The colon is always used to denote a range in Matlab. In this case, 1:10 indicates to Matlab to form a vector starting with the value 1, incrementing in steps of 1 (this is not stated, it is the default), up to 10. If we wanted instead to form the vector w=[1 3 5 7 9] we could use the syntax

w = 1:2:9

which indicates that Matlab should form a vector starting with 1, incrementing by 2, up to 9. Note that we could accomplish the same thing using the command w = 1:2:10.5. Thus, we see that the syntax a:b:c in Matlab represents form a vector starting with a, incrementing by b, up to the number less than or equal to c. Note that none of the numbers a, b or c are required to be integers.

Now that we have vectors, we might have occasion to refer to just one of their entries. This can be done using the index inside parentheses. Using the example vectors above, we can refer to the third entry in v by typing v(3); the result is 3. Likewise we can get the third entry in w: w(3) would be 5. We can set the values of individual entries in vectors in the same way.

w(3) = 293 w = 1 3 293 7 9

We can go further. Suppose we want to change the first, second, and third entries in w. We could make three lines giving e.g. w(1) = 11; w(2) = 22; w(3) = 33;, but we could also give a vector as the index.

w(1:3) = [11 22 33] w = 11 22 33 4 5

Matrices behave in a very similar way, but in this case the indices are ordered pairs of integers. Consider

A = [1 2 3; 4 5 6; 7 8 9] A = 1 2 3 4 5 6 7 8 9

We could accomplish this same assignment using the command A = [1:3;4:6;7:9]. One would not want to do this, but it is instructive to note that we could also do this assignment as A=[[1:3:7]' [2:3:8]' [3:3:9]']. In the last case, we form the matrix as a collection of column vectors. Now we can refer to the 1 in the matrix as A(1,1); the 3 is A(1,3); the 7 is A(3,1). We could extract the 2×2 submatrix in the upper right corner of A using

B = A(1:2,2:3) B = 2 3 5 6

In the same way one can use vector indices to assign entire rows or columns, as in the following.

A(:,3) = [0 0 0]' A = 1 2 0 4 5 0 7 8 0

Note that the notation A(:,3) indicates that the assignment is being done to the entire third column of A. Thus, the above example says: "replace the entire third column of A with the transpose of a zero row vector of dimension 3."

Assignment 6 is posted.

The midterm exam will take place Monday, 16 October. As with all exams, all paper notes may be used, but no electronic devices are permitted. There is a sample exam available.

Department of Mathematics, PO Box 643113, Neill Hall 103, Washington State University, Pullman WA 99164-3113, 509-335-3926, Contact Us
Copyright © 1996-2015 Kevin Cooper