Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Given $\mathbf{A} = \mathbf{B}\mathbf{C}$, with $\mathbf{B} \in \mathbb{R}^{m\times n}$ and $\mathbf{C} \in \mathbb{R}^{n\times p}$. Say we know that the columns of $\mathbf{A}$ are linearly independent. Does this also imply that the columns of $\mathbf{B}$ and $\mathbf{C}$ are linearly independent? How can I go about proving this?

We know $\text{Ker}(\mathbf{BC}) = \left\{\mathbf{0}\right\}$, so $\mathbf{x} = \mathbf{0}$ is the only vector satisfying $$ \mathbf{BCx} = \mathbf{0} $$

From this: $$ \mathbf{BC}(\mathbf{0}) = \mathbf{B}\mathbf(0) = \mathbf{0} $$

but this only shows that $\mathbf{0}$ is in the null space of $\mathbf{B}$, not that it's the only member of the nullspace.

share|improve this question
    
do you have any idea about determinants? –  Praphulla Koushik Oct 14 '13 at 4:51
    
@PraphullaKoushik wouldn't determinants only be helpful if these were square matrices? –  user100704 Oct 14 '13 at 4:53
    
Oh, yes yes.. I did not realize given matrices are not square matrices –  Praphulla Koushik Oct 14 '13 at 4:55

2 Answers 2

up vote 0 down vote accepted

It is not necessarily true that the columns of $B$ are linearly independent. For example, $$ \pmatrix{1 & 0\cr 0 & 1\cr} = \pmatrix{1 & 0 & 0\cr 0 & 1 & 0\cr} \pmatrix{1 & 0\cr 0 & 1\cr 0 & 0\cr}$$ On the other hand, it is true that the columns of $C$ are linearly independent, because ${\rm Ker}(C) \subseteq {\rm Ker}(BC)$.

share|improve this answer
    
Thank you. Can you elaborate on the last statement? I can kind of see how to prove it. Suppose $\mathbf{x} \in \text{Ker}(\mathbf{C})$. $\mathbf{Cx} = \mathbf{0}$ so we have $\mathbf{BCx} = \mathbf{0}$, so x will be in $\text{Ker}(\mathbf{BC})$. But how do we prove there isn't an element of $\Ker(\mathbf{C})$ that isn't in $\text{Ker}(\mathbf{BC})$? By simple contradiction? –  user100704 Oct 14 '13 at 5:15
    
Huh? You just proved that: if $x \in {\rm Ker} C$, then $x \in {\rm Ker}(BC)$. –  Robert Israel Oct 14 '13 at 5:33
    
wow, sorry, total brain fart. thanks. –  user100704 Oct 14 '13 at 5:35

The column vector of $C$ must be linear independent. You might try to show that if $A = BC$ are just mappings and $A$ is injective, then so is $C$. However, $B$ might not be injective.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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