I understood the concept of dimension of a vector space, which is the number of vectors in all basis of a certain vector space $V$. I understand that, for example, the $dim(\mathbb{R}^2)=2$ or $dim(\mathbb{R}^3)=3$, but demonstrate the correctness of these statements is more complicated.

For example, why the dimension of a general vector space $M_{3\times4}$ matrices is $3\times 4=12$?

Apparently, all matrices that form the standard basis for a vector space $M_{3\times4}$ matrices are special matrices: they have all zeros, except in the positions $ij$, where $i$ refers to the rows and $j$ to the columns, where they have $1$.

For example:

$\left[\begin{matrix}1 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \\ \end{matrix}\right]$

Would be one of the matrices of the standard basis of the vector space $M_{3x4}$, where $i=1$ and $j=1$, right?

What do I need to do show to demonstrate it is indeed $12$?


Let $E_{ij}\in M_{3 \times 4}$ be the matrix with $1$ at the $ij$-entry and zeroes everywhere. For example: $$E_{12} = \begin{pmatrix} 0 & 1 & 0 & 0 \\ 0 & 0 & 0 & 0 \\ 0 &0 & 0 & 0\end{pmatrix}.$$

Let $A = (a_{ij}) \in M_{3 \times 4}$. Then we have $A = \sum_{i=1}^3\sum_{j=1}^4a_{ij}E_{ij}$. This shows that $\{E_{ij} \mid 1 \leq i \leq 3,\, 1 \leq j \leq 4\}$ spans $M_{3 \times 4}$, so $\dim (M_{3 \times 4}) \leq 12$. Showing that $\{E_{ij} \mid 1 \leq i \leq 3,\,1 \leq j\leq 4\}$ is linearly independent ensures that $\dim(M_{3 \times 4}) \geq 12$.

So we conclude that $\dim(M_{3 \times 4}) = 12$.

Another way is noticing that $$M_{3 \times 4} \ni (a_{ij}) \mapsto (a_{11},a_{12},\cdots, a_{21},\cdots, a_{31},\cdots, a_{34})\in \Bbb R^{12}$$ is a bijection which preserves vector sum and multiplication by scalar, hence $\dim(M_{3 \times 4}) = \dim(\Bbb R^{12})=12.$

| cite | improve this answer | |
  • $\begingroup$ How would you approach in showing that they are linearly independent? $\endgroup$ – nbro Mar 21 '15 at 3:35
  • $\begingroup$ Using the definition: check that $$\sum_{i=1}^3\sum_{j=1}^4a_{ij}E_{ij} = 0_{3\times 4}\implies a_{ij} = 0,\,\forall\,i,\,\forall\,j.$$ $\endgroup$ – Ivo Terek Mar 21 '15 at 3:38
  • $\begingroup$ Ok, but how would you check? I have done it with vectors, by using $RREF$ (if I am not wrong), but for matrices, how would you proceed? $\endgroup$ – nbro Mar 21 '15 at 3:42
  • $\begingroup$ You don't need RREF to do anything of this. Bear in mind the definition of the $E_{ij}$ and that the $a_{ij}$ in my last comment are only real numbers, suck it up and expand that double summation. It might be useful recalling the definition of linear independence too. $\endgroup$ – Ivo Terek Mar 21 '15 at 3:44
  • $\begingroup$ Expanding the double summation is good for seeing what happens. If you convince yourself that the $ij$-entry of $\sum_{i=1}^3\sum_{j=1}^4a_{ij}E_{ij}$ (remember, this crap is a matrix!) is only $a_{ij}$, then there's nothing else to do. And while we're at it, notice how irrelevant the numbers $3$ and $4$ are, here. Our reasoning goes for $M_{n \times m}$, $\sum_{i=1}^n\sum_{j=1}^ma_{ij}E_{ij} = 0_{n\times m}$, etc. $\endgroup$ – Ivo Terek Mar 21 '15 at 3:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.