1
$\begingroup$

Let $f$ be holomorphic on $\mathbb{C}$. We have $f(z)=\sum_{n=0}^\infty a_nz^n$. Let $g$ be defined on $\mathbb C\setminus\{0\}$ by the Laurent series $g(z)=\sum_{n=0}^\infty \frac{a_n}{z^n}$. If $0$ is an essential singularity of $g$, then the coefficients satisfy $a_n \neq 0$ for infinitely many $n$, so we have an infinite number of terms in the Laurent series. Why does this imply that $f$ is not a polynomial?

$\endgroup$
  • $\begingroup$ Corrected both mistakes. $\endgroup$ – Chris Jun 17 '12 at 21:20
2
$\begingroup$

This comes down to the uniqueness of Taylor coefficients. By assumption, $f(z)=a_0+a_1z+a_2z^2+\cdots$, where the series converges on all of $\mathbb C$. If $f$ were a polynomial, then there would exist an $n\in\mathbb N$ and complex numbers $b_0,b_1,\ldots,b_n$ such that $f(z)=b_0+b_1z+b_2z^2+\cdots+b_nz^n$ for all $z\in\mathbb C$.

But then for $k\in\{0,1,2,\ldots,n\}$, $b_k=a_k=\dfrac{1}{k!}f^{(k)}(0)$, and for $k>n$, $a_k=\dfrac{1}{k!}f^{(k)}(0)=0$. Therefore, if $f$ is a polynomial, then $a_k=0$ for all but finitely many $k$.

| cite | improve this answer | |
$\endgroup$

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.