55
$\begingroup$

Let $A$ and $B$ be two matrices which can be multiplied. Then $$\operatorname{rank}(AB) \leq \operatorname{min}(\operatorname{rank}(A), \operatorname{rank}(B)).$$

I proved $\operatorname{rank}(AB) \leq \operatorname{rank}(B)$ by interpreting $AB$ as a composition of linear maps, observing that $\operatorname{ker}(B) \subseteq \operatorname{ker}(AB)$ and using the kernel-image dimension formula. This also provides, in my opinion, a nice interpretation: if non-stable, under subsequent compositions the kernel can only get bigger, and the image can only get smaller, in a sort of loss of information.

How do you manage $\operatorname{rank}(AB) \leq \operatorname{rank}(A)$? Is there a nice interpretation like the previous one?

$\endgroup$
  • $\begingroup$ Your proof is fine. Furthermore, the same reasoning will get your desired fact. Again rank-nullity will tell you that the dimension of your vector space minus the dimension of the kernel will give you the rank. $\endgroup$ – BBischof Jul 28 '10 at 16:08
29
$\begingroup$

Yes. If you think of A and B as linear maps, then the domain of A is certainly at least as big as the image of B. Thus when we apply A to either of these things, we should get "more stuff" in the former case, as the former is bigger than the latter.

| cite | improve this answer | |
$\endgroup$
  • 2
    $\begingroup$ Thank you. I was so obsessed with the kernel-image dimension formula that I could't recognize this simple fact. $\endgroup$ – user365 Jul 30 '10 at 8:52
20
$\begingroup$

Once you have proved $\operatorname{rank}(AB) \le \operatorname{rank}(A)$, you can obtain the other inequality by using transposition and the fact that it doesn't change the rank (see e.g. this question).

Specifically, letting $C=A^T$ and $D=B^T$, we have that $\operatorname{rank}(DC) \le \operatorname{rank}(D) \implies \operatorname{rank}(C^TD^T)\le \operatorname{rank} (D^T)$, which is $\operatorname{rank}(AB) \le \operatorname{rank}(B)$.

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ Very nice! Thank you. $\endgroup$ – user365 Sep 2 '10 at 10:01
  • $\begingroup$ I am doing this problem now too. Why is proving both inequalities enough? Where does the proof take the min idea into account? $\endgroup$ – MathIsHard Apr 12 '16 at 3:27
  • 1
    $\begingroup$ @MathisHard $x<y \wedge x<z \implies x<\min(y,z)$. $\endgroup$ – stillanoob May 25 '18 at 15:47
10
$\begingroup$

Prove first that if $f:X\to Y$ and $g:Y\to Z$ are functions between finite sets, then $|g(f(X))| \leq \min \{ |f(X)|, |g(Y)| \}.$

Then use the same idea.

| cite | improve this answer | |
$\endgroup$
  • 1
    $\begingroup$ Categorification... :-) $\endgroup$ – Kevin H. Lin Jul 29 '10 at 21:50
  • $\begingroup$ I am not familiar with the 'categorification'. How can one go from this to the rank inequality ? What functor is to be applied ? $\endgroup$ – nicolas Apr 13 '14 at 12:10
  • $\begingroup$ So you're saying that rank of linear map is somehow like image of a function? $\endgroup$ – Mihail Feb 12 '15 at 10:15
  • 2
    $\begingroup$ @Mihail, it is like the size of the image of a function. in fact, if $f$ is a linear map, the rank of $f$ is the dimension of the image of $f$ and the dimension is a measure of the size of a space. $\endgroup$ – Mariano Suárez-Álvarez Feb 12 '15 at 18:22
9
$\begingroup$

Note that $\text{Col}(AB) ⊆ \text{Col}(A)$ since given $y ∈ \text{Col}(AB)$ we can choose $x ∈ F$ and then we have $y = (AB)x = A(Bx) ∈ \text{Col}(A)$.

Since $\text{Col}(AB) ⊆ \text{Col}(A)$, any basis for $\text{Col}(AB)$ can be extended to a basis for $\text{Col}(A)$ and so $\dim \text{Col}(AB) ≤ \dim \text{Col}(A)$, that is $\text{rk}(AB) ≤ \text{rk}(A).$

Note that $\text{Null}(B) ⊆ \text{Null}(AB)$ since given $x ∈ \text{Null}(B)$ we have $(AB)x = A(Bx) = A 0 = 0$ so that $x ∈ \text{Null}(AB)$.

Since $\text{Null}(B) ⊆ \text{Null}(AB)$, as above we have $\dim \text{Null}(B) ≤ \dim \text{Null}(AB)$, that is, $\text{Nullity}(B) ≤\text{Nullity}(AB)$. Thus $\text{rk}(AB) = n − \text{Nullity}(AB) ≤ n − \text{Nullity}(B) = \text{rk}(B).$

| cite | improve this answer | |
$\endgroup$
  • 4
    $\begingroup$ You should use MathJax to format your answer. $\endgroup$ – Michael Albanese Oct 15 '15 at 21:06
  • $\begingroup$ Very nice proof! +1 $\endgroup$ – ZFR Apr 23 at 1:41
5
$\begingroup$

Here is another simple answer. When you multiply a matrix and a vector $Ax$ you end up with a linear combination of the columns of $A$.

$$ Ax = \; x_1\,A_1 \;+\; x_2\,A_2 \;+\; x_3\,A_3 \;+\;\; ...\;\; \\ $$

When we multiply two matrices $AB = C$, we have $AB_i = C_i$, which means that each column of $C$ is a linear combination of the columns of $A$, so $\text{rank}(AB) \leq \text{rank}(A)$. To show that $\text{rank}(AB) \leq \text{rank}(B)$ we follow a similar argument -- when you multiply $x^{\top}B$, you end up with a linear combination of the rows of $B$.

| cite | improve this answer | |
$\endgroup$
3
$\begingroup$

Already you have proved $$rank(AB)\leq rank(B)$$

For other part

rank of $ A=$ dim range $A$

As range $AB \subset $ range $A\implies $ dim range $AB\leq $ dim range $A$ . Hence $$rank(AB)\leq rank (A)$$

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ Hello!! Why does it hold that "range AB $\subset$ range A" ? $\endgroup$ – Mary Star Jan 7 '17 at 10:01
  • $\begingroup$ If $x\in$ range $AB,$ then $x=(AB)y$ for some $y.$ Thus $x=A(By).$ Thus $ x\in$ range $A.$@ Mary Star $\endgroup$ – Black-horse Jan 8 '17 at 17:48
2
$\begingroup$

Let $ m \le n, A \in M_{m\times n}, B\in M_{n\times m}$.

$\mbox{rank } A\le m$ and $\text{rank }B\le m$. (Obvious fact as rank A = dimension of the columnspace of A = dimension of the row space of A.)

Let $E_{n\times n}B$ be the row echelon form of $B$ and let $AE_{m\times m} $ be the column echelon form of $A$. ($E_{n\times n} ,E_{m\times m}$ are elementary matrices.)

We know $\operatorname{rank}(BA)=\operatorname{rank}(E_{n\times n}BA )=\operatorname{rank}(E_{n\times n}BAE_{m\times m} )$.

But $E_{n\times n}BAE_{m\times m} =\begin{pmatrix} L&0\\ 0&0\\ \end{pmatrix},$ where $L$ is an $k\times l$ matrix with $k\le \operatorname{rank}(B),l\le \operatorname{rank}(A)$.

So $\operatorname{rank}(E_{n\times n}BAE_{m\times m} )=\operatorname{rank}\begin{pmatrix} L&0\\ 0&0\\ \end{pmatrix}\le \min\{k,l\}\le \min\{\mbox{rank } A,\mbox{rank }B\}.$

| cite | improve this answer | |
$\endgroup$
0
$\begingroup$

Another way of showing that $\text{Rank} (AB) \leq \text{Rank}(B)$ without using rank-nullity: Note that if $v_1,\dots,v_n$ is a basis of $\text{Range} B$, then $Av_1,\dots,Av_n$ is a spanning list of $\text{Range} AB$.

| cite | improve this answer | |
$\endgroup$