# Multiply a matrix by a matrix ¶

Go back

If the number of columns of a matrix $M1_{n1, \ {\color{red}c}}$ is the same as the number of lines of a matrix $M2_{\ {\color{red}c}, p2}$ then you can multiply $M1$ by $M2$

• the result is a matrix $M$: $M1_{n1,c} * M2_{c, p2} = M_{n1,p2}$.
• $M1 * M2$ DOES NOT mean that we can do $M2 * M1$.

## Example ¶

Here is an example of what you need to do

$\begin{split}\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ M1\ \begin{pmatrix} \color{blue}{1} & 4 \\ \color{blue}{2} & 5 \\ \color{blue}{3} & 6 \end{pmatrix}\end{split}$ $\begin{split}M2 \begin{pmatrix} \color{red}{9} & \color{red}{8} & \color{red}{7} \\ 6 & 5 & 4 \end{pmatrix} \ M \begin{pmatrix} \color{red}{9}*\color{blue}{1}+\color{red}{8}*\color{blue}{2}+\color{red}{7}*\color{blue}{3}=46 & 9*4+8*5+7*6=118 \\ 6*1+5*2+4*3=28 & 6*4+5*5+4*6=73 \end{pmatrix}\end{split}$

As for an explanation

• we usually put your matrix in a reverse L
• on top, the first matrix
• on the left side, the second matrix
• and in the corner, the resulting matrix
• for the value $a_{\color{red}{1}\color{blue}{1}}$ of the resulting matrix, we are doing
$\sum_{i=1}^n a_{{\color{red}1},i} * b_{i, \color{blue}{1}}$

## Code in R ¶

# matrix 3x3, with (1,2,3\\4,5,6\\7,8,9)
A <- matrix(1:9, 3, 3, byrow = TRUE)
# matrix of 1
B <- matrix(rep(1, length = 9), 3, 3, byrow = TRUE)

A %*% B
#       [,1] [,2] [,3]
# [1,]    6    6    6
# [2,]   15   15   15
# [3,]   24   24   24