Evaluating the matrix polynomial

83 Views Asked by At

I have a question that goes as follows:

Let $\alpha = \frac{\pi}{5}$ and $A= \begin{bmatrix} \cos \alpha & \sin \alpha \\ -\sin \alpha & \cos \alpha \\ \end{bmatrix} $ and let $B=A^4-A^3+A^2-A$

Then which of the following is true? (i)B is singular (ii) B is non singular (iii) B is symmetric (iv) The determinant $|B| =1$

There may be multiple correct options.

Now here's what I did, I evaluated $A^2$ and $A^3$ and then it is pretty easy to conclude by induction that $$A^n = \begin{bmatrix} \cos n\alpha & \sin n\alpha \\ -\sin n\alpha & \cos n\alpha \\ \end{bmatrix}$$

From this, I tried to find out $B$ by putting the values of $A^4, A^3, A^2, A$ and then applied the transformation formulae $\bigg($like $\sin C + \sin D=$ $2\sin\bigg( \frac{C+D}{2} \bigg)\cos\bigg( \frac{C-D}{2}$ $\bigg)$, $...$ etc$\bigg)$

By simplifying, and using the fact that $\alpha = \frac{\pi}{5}$ I got the matrix B as follows:

$$B = \begin{bmatrix} -4\cos \alpha \sin \frac{\alpha}{2} & 0 \\ 0 & -4\cos \alpha \sin \frac{\alpha}{2} \\ \end{bmatrix}$$

So, clearly, the matrix is symmetric, which makes (iii) true. But, how do I check whether the matrix is non-singular or not, and how do I check whether $|B|=1$? Using the manipulations that I performed, I seem to not be able to check those options, any help would be nice.

The answer is given as (ii), (iii), (iv) are true.

1

There are 1 best solutions below

11
On BEST ANSWER

One way to think of this is that you are essentially computing a geometric sum associated to $-A$. As usual, we have that $(-A - 1)B = -A^5 + A = 1 + A$. Since $-A - 1$ is non-singular (as you can compute by finding the determinant, by computing its eigenvalues, or just by explicitly writing down the inverse of a $2\times2$ matrix), we see that $B = -1$. Sure enough, this is symmetric and has determinant $1$.

It may help in this case to think that we are essentially computing with complex numbers; the ring of complex numbers $\mathbb C$ embeds in the ring $\operatorname M_{2\times2}(\mathbb R)$ of $2\times2$ real matrices by $a + b i \mapsto \begin{pmatrix} a & b \\ -b & a \end{pmatrix}$ for all $a, b \in \mathbb R$. In particular, you are really computing a sum inside $\mathbb C$ involving $z = \cos(\alpha) + i\sin(\alpha) = e^{i\alpha}$. In this context, the determinant on $\operatorname M_{2\times2}(\mathbb R)$ pulls back to the squared-norm on $\mathbb C$, in the sense that $$ \det \begin{pmatrix} a & b \\ -b & a \end{pmatrix} = a^2 + b^2 = \lvert a + b i\rvert^2 $$ for all $a, b \in \mathbb R$; and the transpose on $\operatorname M_{2\times 2}(\mathbb R)$ pulls back to complex conjugation on $\mathbb C$, in the sense that $$ a + b i \mapsto \begin{pmatrix} a & b \\ -b & a \end{pmatrix} \quad\text{and}\quad \overline{a + b i} = a - b i \mapsto \begin{pmatrix} a & -b \\ b & a \end{pmatrix} = \begin{pmatrix} a & b \\ -b & a \end{pmatrix}^{\mathsf T} $$ for all $a, b \in \mathbb R$ (and remember that $\overline{e^{i\theta}}$ equals $e^{-i\theta}$ for all $\theta \in \mathbb R$); so asking whether the alternating sum of matrices is symmetric, respectively has determinant $1$, is the same as asking whether the corresponding alternating sum of complex numbers is real, respectively lies on the unit circle centred at the origin. Finally, since $\mathbb C$ is a field, every non-$0$ matrix in the image of $\mathbb C$ is invertible (with the inverse of the image of $z = a + b i$ being the image of $z^{-1} = \frac a{a^2 + b^2} - \frac b{a^2 + b^2}i$ for all $(a, b) \in \mathbb R^2 \setminus \{(0, 0)\}$); this is yet another way to see that $-A - I_2$ is non-singular.