Suppose that $f$ is holomorphic inside and on $\gamma(0;1)$, the circle centered at $0$ of radius $1$, with Taylor expansion $\sum_{n=0}^{\infty} c_nz^n$. Given that $f$ has $m$ zeros inside $\gamma(0;1)$, prove that $$\min \{ |f(z)| : |z|=1 \} \leq |c_0| + |c_1|+ \dots + |c_m|$$

I really don't know how to start. From the hypothesis we know that

$$\frac{1}{2\pi i}\int_{\gamma(0;1)}\frac{f'(z)}{f(z)} dz =m$$ but I'm not sure if that fact is useful at all. Maybe I should exploit the series expansion of $f(z)$, but I haven't come up with anything. Any help will be highly appreciated, and thanks in advance!

  • $\begingroup$ What does $\gamma(0,1)$ stand for? Is it the open disk of unit radius centred at the origin? $\endgroup$ – Arian Jul 29 '16 at 17:21
  • $\begingroup$ @Arian It is the circle centerd at zero with radius $1$, I'll edit the question to make that clearer. $\endgroup$ – user313212 Jul 29 '16 at 17:27

I thank very much @See-Woo Lee who has indicated the right way to prove the issue, i.e., by contradiction.

Let $f(z)=p(z)+q(z)$ with $\begin{cases}p(z)&=&\sum_{n=0}^{m}c_nz^n\\q(z)&=&\sum_{n=m+1}^{\infty} c_nz^n \end{cases}$.

Let $U$ be the unit disk and $\partial U$ be the unit circle.

Let us prove the result by contradiction.

Let us assume the contrary of what has to be proven, i.e., that

$$\forall z \in \partial U, \ |f(z)| > |c_0| + |c_1|+ \dots + |c_m|$$

thus, a fortiori

$$\forall z \in \partial U, \ |f(z)| > |c_0 + c_1 z+ \dots + c_m z^m|=|p(z)|=|f(z)-q(z)|$$

Using Rouché's theorem, $f(z)$ and $f(z)-q(z)=p(z)$ have the same number of zeros inside the unit disk $U$, i.e., $m$ zeros.

But $q(z)=z^{m+1}r(z)$ where $r(z)$ is a holomorphic function; therefore $q(z)$ would have at least $m+1$ zeros in $U$, a contradiction with the previous sentence.

  • 2
    $\begingroup$ can you expand the reasoning ? $\endgroup$ – Hamza Jul 29 '16 at 17:36
  • 1
    $\begingroup$ So I should proceed by comtradiction? Assume that $\vert \sum_{n=m+1}^{\infty} c_nz^n \vert < |f(z)|$. Then $g(z)$ has $m$ zeros inside the contour $\gamma(0;1)$, but I can't see any contradiction there. $\endgroup$ – user313212 Jul 29 '16 at 19:52
  • $\begingroup$ I have provided a thorough proof now ; yes, you were on the good track by working by contradiction. $\endgroup$ – Jean Marie Jul 29 '16 at 22:01
  • $\begingroup$ Thanks, @Hamza, for the correction of "polynomial" to "holomorphic". $\endgroup$ – Jean Marie Jul 30 '16 at 0:10
  • $\begingroup$ excuse me because i do it without telling you $\endgroup$ – Hamza Jul 30 '16 at 0:18

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.