Prove that $|\mathbf{A}+\mathbf{B}|=\mathbf{0}$ As per following condition

91 Views Asked by At

If $\mathbf{A}$ and $\mathbf{B}$ are real orthogonal matrices of the same order and $|\mathbf{B}|+|\mathbf{A}|=\mathbf{0}$ Prove that $|\mathbf{A}+\mathbf{B}|=\mathbf{0}$


My Approach:-

$|\mathrm{A}|+|\mathrm{B}|=0$

$\Rightarrow|A|=-|B|$

$|\mathrm{A}| \cdot|\mathrm{B}|=-1 \quad\left[\because|\mathrm{B}|=\left|B^{-1}\right| \text {as they are orthogonal }\right]$

Let, $C=A\left(A^{T}+B^{T}\right) B$

$\Rightarrow|C|=\left|A A^{T} B+A B^{T} B\right|=|B+A| \ldots \ldots \ldots$ (i)

And $|\mathrm{C}|=|\mathrm{A}|\left|A^{T}+B^{T}\right||B|=-\left|A^{T}+B^{T}\right|$

$\Rightarrow-\left|(A+B)^{T}\right|=-|A+B| \ldots \ldots \ldots \ldots$ (ii)

$|A+B|=-|A+B|$

$\Rightarrow 2|A+B|=0$

$\boxed{\Rightarrow|A+B|=0}$

I am looking for another short approach.Any alternate solution would be greatly appreciated

1

There are 1 best solutions below

4
On BEST ANSWER

$\def\norm[1]{\lVert #1 \rVert}$Here is a proof that I think gives more intuition. Let $X$ be a real orthogonal matrix.

Lemma 1. $|\lambda(X)| = 1$

Proof. $Xu=\lambda u \Rightarrow |\lambda|^2 \lVert u \rVert^2 = \lVert Xu \rVert^2 = u^* X^TXu = \lVert u \rVert^2$

Lemma 2. If $|X|=-1$ then $X$ has an odd number of $-1$ eigenvalue.

Proof. If $\lambda$ is complex then $\bar{\lambda}$ is also an eigenvalue of $X$. So, $-1=|X|=\prod_i \lambda_i = (-1)^k$ since multiplication of other eigenvalues results in $1$.

Answer. $|A|+|B|=0 \Rightarrow |A^TB| = -1$, which means $A^TB$ (which is also orthogonal) has at least 1 eigenvalue of $-1$, so $|I+A^TB|=0$ and consequently $|A+B|=0$.