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.

Let $V=\mathbb{R}^{n}$ and $T\,:V\to V$ be defined by $Tv=Av$ where $A\in M_{n}(\mathbb{R})$ is an orthogonal matrix.

My lecture wrote that if $W\subset V$ is a subspace of $V$ then if $W$ is $A$ invariant then $W^{\perp}$ is also $A$ invariant.

What I do know is that if $W$ is $A$ invariant then $W^{\perp}$ is also $A^*=A^{t}$ invariant, but I could not deduce from this that it is also $A$ invariant.

Is this 'fact' true ? I couldn't prove it (I tried writing a proof similar to the case I know, using inner products and failed), help is appreciated!

share|improve this question
1  
Remember that $\langle Au, Av \rangle = \langle u,v \rangle$ whenever $u,v$ are column vectors in $\mathb{R}^n}$ and $\langle,\rangle$ is the usual inner (dot) product. –  Geoff Robinson Jul 14 '12 at 12:05
add comment

2 Answers

up vote 3 down vote accepted

If $$\forall\,w\in W\,\,,\,Aw=w'\in W\Longrightarrow A^{-1}w'=A^{-1}Aw=w\in W$$

The above is enough since we know $\,A\,$ is bijective (why?), so for any $\,w'\in W\,$ there always exists $\,w\in W\,\,\,s.t.\,\,\,Aw=w'\,$

Thus, $\,W\,\,is\,\,A-\,$ invariant iff it is $\,A^{-1}-\,$ invariant, and now use Marc's answer.

share|improve this answer
add comment

You need to use that $A$ is orthogonal, i.e., $A^t=A^{-1}$. Can you show that a subspace is $A$-invariant if and only if it is $A^{-1}$-invariant?

share|improve this answer
    
I tried to show this as well and could not do it, I tried something like $A^{-1}w=A^{-1}(A^{-1}Aw)=A^{-1}A^{-1}w'$ and it doesn't help me... –  Belgi Jul 14 '12 at 12:01
    
Try writing $V=A\cdot V$ and applying $A^{-1}$ to both sides... –  Marc van Leeuwen Jul 14 '12 at 13:02
add comment

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.