Proving that the limit of a function doesn't exist at an essential singularity

725 Views Asked by At

Assume $f$ is analytic on a deleted neighborhood $B'(0;a)$. Prove that the limit of the function as $z$ approaches $0$ exists (possibly infinite) if, and only if there exists an integer $n$ and a function $g(z)$ analytic on $B(0;a)$, with $g(0)$ not zero such that $f(z)=z^ng(z)$ in $B'(0;a)$

The "if" direction is simple but I am struggling with the other direction. The "only if" part is the same as proving that if the principle part of Laurent series expansion for the function around $0$ is an infinite sum, the limit doesn't exist which is the same as proving that it depends on the direction in which we approach $0$. I tried to exhibit two different limits but failed mainly due to the fact that I couldn't deal with the arbitrary coefficients of the series (since this should hold for any function with an essential singularity). This is a problem in Apostol's mathematical analysis and he doesn't mention the Casorati–Weierstrass theorem so I suppose one should be able to solve the problem without depending on it.

2

There are 2 best solutions below

0
On BEST ANSWER

If the limit $\lim_{z\to0}f(z)$ exists in $\mathbb C$, then $a$ is a removable singularity and you can apply Riemann's theorem on removable singularities.

Otherwise, $\lim_{z\to0}f(z)=\infty$. So, there is some $r\in(0,a)$ such that $f$ has no zeros in $B(0;r)$. Let$$\begin{array}{rccc}\varphi\colon&B(0;r)&\longrightarrow&\mathbb C\\&z&\mapsto&\begin{cases}\frac1{f(z)}&\text{ if }z\neq0\\0&\text{ if }z=0.\end{cases}\end{array}$$Then $\varphi$ is analytic, $0$ is a removable singularity of $\varphi$ and you can apply Riemann's theorem to it.

0
On

If the limit is finite then $f$ has a removable singularity at $0$. If $n$ is the order of zero of $f$ at $0$ (which may be $0$) then we can write $f(z)=z^{n}g(z)$ with $g$ analytic with no zeros in some disk around $0$. If the limit of $f$ is $\infty$ then $f$ has a pole at $0$. In this case let $-n$ be the order of the pole at $0$. See Rudin's RCA for details.