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.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $V$ be a finite-dimensional inner product space and let $x,y\in V$ be nonzero vectors. If there is a self-adjoint operator $A:V\rightarrow V$ such that $A(x)=y$ and $\langle A(v),v\rangle\geq0$ for all $v\in V$, then $\langle x,y\rangle>0$.

I think we can conclude the following inequality $$\langle x,y \rangle=\langle x,A(x) \rangle=\langle A^*(x),x \rangle=\langle A(x),x\rangle\geq 0$$

but I'm not able to show that strict inequality holds (or, in other words, that equality is not possible). Can someone give me a hint?

share|cite|improve this question
I added the operator theory tag since the result is true more generally (and can be proved the same way) on a Hilbert space, with $A$ a bounded self-adjoint positive operator. – 1015 Jul 1 '13 at 1:04
up vote 1 down vote accepted

Assume for a contradiction that $(x,y)=0$. Then take an orthonormal basis $(e_j)$ of $V$ starting with $$ e_1=\frac{x}{\|x\|}\qquad e_2=\frac{y}{\|y\|}. $$ The matrix $M$ of $A$ in this orthonormal basis must be symmetric semidefinite positive, and it looks like $$ M=\pmatrix{0&t&0\\ t&*&*\\ 0&*&*}\qquad t=\frac{\|y\|}{\|x\|}>0 $$ by blocks, where the first and the second row are the actual first and second rows of $M$, the remaining blocks being of the ad hoc size.

Can you see a contradiction?

Consider the restriction of the quadratic form $(Ax,x)$ to the span of $\{e_1,e_2\}$, whose matrix is the upper-left $2\times 2$ block of $M$. Since the restriction should be semidefinite positive as well, its eigenvalues, whence its determinant $-t^2<0$, must be nonnegative. Contradiction.

share|cite|improve this answer

Your Answer


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.