# Is the Set of All Orthogonal Matrices a Vector Space?

## Problem 611

An $n\times n$ matrix $A$ is called **orthogonal** if $A^{\trans}A=I$.

Let $V$ be the vector space of all real $2\times 2$ matrices.

Consider the subset

\[W:=\{A\in V \mid \text{$A$ is an orthogonal matrix}\}.\]
Prove or disprove that $W$ is a subspace of $V$.

Contents

## Solution.

We claim that $W$ is not a subspace of $V$.

One way to see that $W$ is not a subspace of $V$ is to note that the zero vector $O$ in $V$, which is the $n\times n$ zero matrix, is not in $W$ as we have $O^{\trans}O=O\neq I$.

Thus, $W$ is not a subspace of $V$.

### Another approach

You may also show that scalar multiplication (or addition) is not closed in $W$.

For example, the identity matrix $I$ is orthogonal as $I^{\trans}I=I$, and thus $I$ is an element in $W$.

However, the scalar product $2I$ is not orthogonal since

\[(2I)^{\trans}(2I)=4I\neq I.\]

Add to solve later