If $a \in \mathbb{R}$, evaluate $$ \lim_{n \to \infty}\left(\begin{matrix} 1&\frac{a}{n}\\\frac{-a}{n}&1\end{matrix}\right)^{n}$$

My attempt: Let $$A = \left(\begin{matrix} 0&a\\-a&0\end{matrix}\right) = -a\left(\begin{matrix} \cos(\frac{\pi}{2})&-\sin(\frac{\pi}{2})\\\sin(\frac{\pi}{2})&\cos(\frac{\pi}{2})\end{matrix}\right)$$ so that $$A^k = (-a)^k \left(\begin{matrix} \cos(\frac{k\pi}{2})&-\sin(\frac{k\pi}{2})\\\sin(\frac{k\pi}{2})&\cos(\frac{k\pi}{2})\end{matrix}\right)$$

Thus, \begin{align}\displaystyle \lim_{n \to \infty}\left(\begin{matrix} 1&\dfrac{a}{n}\\\dfrac{-a}{n}&1\end{matrix}\right)^{n} &=\displaystyle \lim_{n \to \infty} \left(I+\dfrac{A}{n}\right)^n =e^A=\displaystyle \sum_{k=0}^{\infty}\dfrac{A^k}{k!}\\&= \sum_{k=0}^{\infty} \left(\begin{matrix} \dfrac{(-a)^k\cos(\frac{k\pi}{2})}{k!}&-\dfrac{(-a)^k\sin(\frac{k\pi}{2})}{k!}\\\dfrac{(-a)^k\sin(\frac{k\pi}{2})}{k!}&\dfrac{(-a)^k\cos(\frac{k\pi}{2})}{k!}\end{matrix}\right) \end{align}

and since $\displaystyle \sum_{k=0}^{\infty}\dfrac{(-a)^k\cos(\frac{k\pi}{2})}{k!}=1+0-\dfrac{a^2}{2!}+0+\dfrac{a^4}{4!}+\cdots= \cos a$ and

$\displaystyle \sum_{k=0}^{\infty}\dfrac{(-a)^k\sin(\frac{k\pi}{2})}{k!}=0-a+0+\dfrac{a^3}{3!}+0-\dfrac{a^5}{5!}+\cdots= -\sin a$ therefore the required answer is

$\left(\begin{matrix} \cos a&\sin a\\-\sin a&\cos a\end{matrix}\right).$

However the above answer does not match the choices provided which are $I, 0$ and none of the above. So my question is: Is my answer correct?

  • 2
    $\begingroup$ your result is correct. $\endgroup$ – julian Nov 18 '17 at 14:55

Your answer is correct as well. but there is a short way to reach it. Merely by putting ourselves in the complex plan where we identify

$$ 1 \equiv \left(\begin{matrix} 1& 0\\0&1 \end{matrix}\right)~~~\text{and}~~~~ i \equiv \left(\begin{matrix} 0& 1\\-1&0 \end{matrix}\right).$$

Thus, $$\begin{align}\displaystyle \lim_{n \to \infty}\left(\begin{matrix} 1&\dfrac{a}{n}\\\dfrac{-a}{n}&1\end{matrix}\right)^{n} &= \displaystyle \lim_{n \to \infty}\color{blue}{\left(1+\dfrac{ai}{n}\right)^n} \\&=\color{red}{e^{ai} = \cos a+i\sin a} \\&= \left(\begin{matrix} \cos a&\sin a\\-\sin a&\cos a\end{matrix}\right).\end{align}$$

Given that, for any $z\in \Bbb C$ we have, $\lim\limits_{n \to \infty}\left(1+\dfrac{z}{n}\right)^n =e^{z} $ See here also here,

  • 1
    $\begingroup$ Thanks for the elegant answer, I should of thought of using the complex plane $\endgroup$ – Aryaman Jal Nov 18 '17 at 15:15
  • $\begingroup$ Yes I had the same problem few year ago where I had to solve some linear ODE which reduce to the compact form $z'= az$ $\endgroup$ – Guy Fsone Nov 18 '17 at 15:23
  • 3
    $\begingroup$ What an answer!! An artwork! $\endgroup$ – Shashi Nov 18 '17 at 16:42
  • 1
    $\begingroup$ @Shashi you might fine mor application here: math.stackexchange.com/questions/164422/… $\endgroup$ – Guy Fsone Dec 1 '17 at 19:21

The best approach is to diagonalize your matrix. Both $(1,i)$ and $(1,-i)$ are eigenvectors. So, let$$T=\begin{pmatrix}1&1\\i&-i\end{pmatrix}.$$Then$$T^{-1}.\begin{pmatrix}1&\frac an\\-\frac an&1\end{pmatrix}.T=\begin{pmatrix}1+\frac ani&0\\0&1-\frac 1ni\end{pmatrix}.$$Therefore$$T^{-1}.\begin{pmatrix}1&\frac an\\-\frac an&1\end{pmatrix}^n.T=\begin{pmatrix}1+\frac ani&0\\0&1-\frac 1ni\end{pmatrix}^n=\begin{pmatrix}\left(1+\frac{ai}n\right)^n&0\\0&\left(1-\frac{ai}n\right)^n\end{pmatrix}$$ and so$$\lim_{n\to\infty}T^{-1}.\begin{pmatrix}1&\frac an\\-\frac an&1\end{pmatrix}^n.T=\begin{pmatrix}e^{ai}&0\\0&e^{-ai}\end{pmatrix}.$$So$$\lim_{n\to\infty}\begin{pmatrix}1&\frac an\\-\frac an&1\end{pmatrix}^n=T.\begin{pmatrix}e^{ai}&0\\0&e^{-ai}\end{pmatrix}.T^{-1}=\begin{pmatrix}\cos a&\sin a\\-\sin a&\cos a\end{pmatrix}.$$


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.