Existence of an Analytic Function on the Unit Disc for $f(1/n)$

1.4k Views Asked by At

I'm reviewing some old exam questions for an upcoming exam and I'm stumped on the following question.

"Does there exist an analytic function on the unit disc with the property that $f(\frac{1} {n}) = \frac{2+n} {3+n^2}$ for $n=2, 3, 4,... $ Give an example or prove it does not exist."

My proof so far is the following: Assume $f$ is analytic. Then as $f$ must be continuous, we define $f(0) = 0$ and can express $f = \sum a_nz^n$.

Let $N$ be the first index such that $a_n \neq 0$. Then $f = a_Nz^N + a_{N+1}z^{N+1} + \ldots = az^Ng(z)$ and where $g(z)$ is analytic and $g(0) = 1$.

Then $|\frac{2+n} {3+n^2}| = |f(\frac{1} {n})| = |a \frac{1} {n^N} g(\frac{1} {n})|$, so $|g(\frac{1} {n})| = |\frac{n^N (2+n)} {a(3+n^2)}|$.

The above expression then tells us that $N = 1$ as otherwise the right hand side would go to infinity as $n$ does. In addition, we necessarily have that $|a| = 1$.

Then $f(\frac{1} {n}) = \frac{a} {n} g(\frac{1} {n})$, so we can find that $g(\frac{1} {n}) = \frac{n^2 + 2n} {an^2 + 3a}$.

As $n \to \infty$, the right hand side goes to $\frac{1} {a}$. However this is not necessarily a contradiction, so I'm stuck. Any advice on how to continue on or advice on a different method would be much appreciated.

1

There are 1 best solutions below

5
On

Based on identity theorem, the function $g(z)=\dfrac{2z^2+2z}{3z^2+4}$ agrees with $f(z)$ in infinite points with a limit point so $f(z)=g(z)$ on $D$.