Let $f\colon B \to \mathbb{C}$ be holomorphic where $B$ is the unit disk in the complex plane centered at the origin. For $0<r<1$, let $M_r=\sup\{|f(z)|:|z|=r\}$.

Show that $|f(0)|≤M_r$ $ \forall r \in (0,1)$.

Now suppose further that $|f(0)|≥|f(z)| \forall z\in B$. Show that $|f(z)|$ is constant on $B$,

For the first bit I use Cauchy's integral formula and got: $$\left| \frac{1}{2\pi i} \int_{|z|=r}^{} \frac{f(z)}{z} \right| = \left|f(0)\right|\leq \frac{1}{2\pi} \int_{|z|=r}^{} \left|\frac{f(z)}{z} \right| $$ Now I'm not sure how to implement the $M_r$

Need help with the second part...

  • $\begingroup$ The following is proper typesetting: $M_r=\sup\{|f(z)|:|z|=r\}$. I edited accordingly. ${}\qquad{}$ $\endgroup$ – Michael Hardy Dec 4 '15 at 1:31
  • $\begingroup$ Thanks is it ok to use | | or is $\left | \right |$ better? $\endgroup$ – Jacob Dec 4 '15 at 1:33
  • $\begingroup$ For something like $|1+f(z)|$ I just write |1+f(z)|. For something like $\displaystyle \left| 1 + \frac{f(z)} z\right|$ I use \left 1 + \frac{f(z)} z\right| because it you omit \left and \right then what you see is $\displaystyle | 1 + \frac{f(z)} z |$. ${}\qquad{}$ $\endgroup$ – Michael Hardy Dec 4 '15 at 2:00

For the first part, you can conclude using the "ML-inequality" (you may know it under a different name):

$$ |f(0)| = \Big| \frac{1}{2\pi i} \int_{|z|=r} \frac{f(z)}{z}\,dz \Big| \le \frac{1}{2\pi} \cdot 2\pi r \cdot \sup_{|z|=r} \Big|\frac{f(z)}{z} \Big| \le r \cdot \frac{M_r}{r} = M_r. $$

  • $\begingroup$ Thanks I know it as the estimation lemma. For the second part I have $0≤|f(z)|≤M_r$. Not sure what to do now. $\endgroup$ – Jacob Dec 3 '15 at 23:50
  • $\begingroup$ I also need help on this, Could you give a hint for the second part $\endgroup$ – George Dec 6 '15 at 12:04

$$ \underbrace{\,\,\left| \frac{1}{2\pi i} \int_{|z|=r} \frac{f(z)}{z} \, dz \right| = |f(0)|\leq \frac{1}{2\pi} \int_{|z|=r} \left|\frac{f(z)}{z} \, dz \right|\,\,}_{\huge \text{?}\vphantom{\displaystyle\int}} $$ Instead of the above, I'd write $$ |f(0)| = \left| \frac{1}{2\pi i} \int_{|z|=r} \frac{f(z)}{z} \, dz \right| \leq \frac{1}{2\pi} \int_{|z|=r} \left|\frac{f(z)}{z} \, dz \right|. $$ The point is we know $\text{“}=\text{''}$ is true because of Cauchy's formula, and we know $\text{“}\le\text{''}$ is true for a different reason, and we initially put $\text{“}=\text{''}$ and $\text{“}\le\text{''}$ only where we already know they're true rather than where we're trying to prove they're true. Then we can say we've proved $$ |f(0)| \leq \frac{1}{2\pi} \int_{|z|=r} \left|\frac{f(z)}{z} \, dz \right|. $$ We can go on to say $$ \frac{1}{2\pi} \int_{|z|=r} \left|\frac{f(z)}{z} \, dz \right| = \frac 1 {2\pi r} \int_{|z|=r} |f(z)| \, |dz|. $$ What does it mean to write $|dz|$ instead of $dz$ (which you omitted)? It means we're integrating with respect to arc length.

That last integral is the average value of $|f(z)|$ on the circle of radius $r$ centered at $0$. So what we're trying to show is that the average value is less than or equal to the largest value. We can say $$ \int_{|z|=r} |f(z)| \, |dz| \le \int_{|z|=r} M_r \, |dz| = 2\pi r M_r. $$

Notice that it would be grave error to write $\displaystyle\int_{|z|=r} M_r \,dz$ instead of $\displaystyle\int_{|z|=r} M_r \, |dz|$, since that integral (with $dz$ instead of $|dz|$) evaluates to $0$ rather than to $2\pi rM_r \vphantom{\displaystyle\int}$.

  • $\begingroup$ Sorry I forgot the dz. What about the second part. I have that |f(z)| is bounded $\endgroup$ – Jacob Dec 4 '15 at 1:55
  • $\begingroup$ Do I need Loiuvilles theorem or something else? $\endgroup$ – Jacob Dec 4 '15 at 2:15
  • $\begingroup$ The value of a harmonic function, including any holomorphic function, at the center of a disk is the average of its values on the boundary of the disk. That will do it. But I'm not sure I want to write a proof of that from scratch right now. $\endgroup$ – Michael Hardy Dec 4 '15 at 2:39
  • $\begingroup$ I also need help with the second part could you please expand on what you mean $\endgroup$ – George Dec 6 '15 at 16:04

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.