How to multiply matrices
Before I show you how to multiply matrices, let's see how to multiply a matrix by a scalar or a real number.
For example, to multiply 4 by a 2x2 matrix, just multiply 4 by every element in the matrix
This technique works well if you don't want to write down the matrix 4 times
Instead of writing down the matrix above 4 times, it is better to do 4 ×


This will work with any matrix. It does not matter what the dimension is, just multiply the scalar by every element in the matrix
Now here is the technique to
multiply matrices
Multiply the elements of each row of the first matrix by the elements of each column of the second matrix. Then, add the products.
Does it make sense? Probably not! Let's illustrate how to multiply matrices with a 2x2 matrix. Once you understand how to do multiplication with a 2x2 matrix, you can do it with matrices of any dimension.

×


=


2 × 6 + 1 × 7 = 12 + 7 = 19


×


=


2 × 5 + 1 × 1 = 10 + 1 = 9


×


=


0 × 6 + 4 × 7 = 0 + 28 = 28


×


=


0 × 5 + 4 × 1 = 0 + 4 = 4

Observation #1:
Did you notice this pattern to follow when you multiply matrices?
row #1 × column #1: answer goes in row #1 column #1
row #1 × column #2: answer goes in row #1 column #2
row #2 × column #1: answer goes in row #2 column #1
row #2 × column #2: answer goes in row #2 column #2
Observation #2:
Matrix on the left (call it matrix A): number of elements in first row determine the number of columns
Matrix on the right (call it matrix B): number of elements in first column determine the number of rows
Since you are multiplying every element in first row by every element in first column, multiplication will not be possible if the number of columns of matrix A is not equal to the number of rows of matrix B
The following multiplication is therefore not possible. Call the matrix on the left A and the matrix on the right B
After you multiply 2 by 6, you got no number to multiply 7 by. It does not work as already stated because the number of columns of matrix A is not equal to the number of rows of matrix B
The product of 2 matrices A and B exists only if the number of columns of A is equal to the number of rows of B
For example, the following matrices can be multiplied. See in green how the number of columns is equal to the number of rows
2x
2 and
2x3
3x
1 and
1x4
4x
3 and
3x1
2x
5 and
5x2
1x
3 and
3x1
The following matrices
cannot be multiplied. See in red that the number of columns of matrix A is not equal to the number of rows of matrix B
2x
2 and
3x3
3x
2 and
1x4
4x
3 and
2x2
2x
5 and
2x5
1x
3 and
1x5
Dimensions of a product matrix
It is easy to find the dimension of a matrix. Let's take a look again at the matrices below we said could be multiplied
2x
2 and
2x3. Dimension is 2x3
3x
1 and
1x4. Dimension is 3x4
4x
3 and
3x1. Dimension is 4x1
2x
5 and
5x2. Dimension is 2x2
1x
3 and
3x1. Dimension is 1x1
How did we get the dimension?
The number of rows of the
product matrix is the number of rows of the matrix on the left
The number of columns of the
product matrix is the number of columns of the matrix on the right
Call the matrix on the left A and the matrix on the right B. Looking at A and B, we can confidently say a couple of things
The dimensions of A is 3x2 and the dimensions of B is 2x4
This means that we can find the product and the dimension of the product is 3x4
There will be
12 multiplications
First row times first column: 1 × 4 + 4 × 0 = 4 + 0 = 4
First row times second column: 1 × 1 + 4 × 1 = 1 + 4 = 5
First row times third column: 1 × 2 + 4 × 1 = 2 + 4 = 2
First row times fourth column: 1 × 1 + 4 × 3 = 1 + 12 = 13
Second row times first column: 0 × 4 + 1 × 0 = 0 + 0 = 0
Second row times second column: 0 × 1 + 1 × 1 = 0 + 1 = 1
Second row times third column: 0 × 2 + 1 × 1 = 0 + 1 = 1
Seecond row times fourth column: 0 × 1 + 1 × 3 = 0 + 3 = 3
Third row times first column: 1 × 4 + 0 × 0 = 4 + 0 = 4
Third row times second column: 1 × 1 + 0 × 1 = 1 + 0 = 1
Third row times third column: 1 × 2 + 0 × 1 = 2 + 0 = 2
Third row times fourth column: 1 × 1 + 0 × 3 = 1 + 0 = 1
We get

4

5

2

13

0

1

1

3

4

1

2

1

