Revision as of 19:08, 15 February 2014

Homework 4 collaboration area

This is the place.

Exercise 1

For what values of $ z $ is the series

$ \sum_{n=0}^{\infty}\left(\frac{z}{1+z}\right)^{n} $

convergent? Same question for

$ \sum_{n=0}^{\infty}\frac{z^{n}}{1+z^{2n}}. $

Discussion

-The first series yields readily to an application of Hadamard's formula and a subsequent analysis of those values $ z $ for which the limsup is equal to $ 1 $. The second series seems a little more difficult. Neither Hadamard nor the ratio test seem to be very easy to work out. Can we recognize these terms as the derivative of another series? We know that the derivative of a power series will have the same radius of convergence as the original series. Should we try to decompose the series somehow? For example, we can write

$ \sum_{n=0}^{\infty}\frac{z^{n}}{1+z^{2n}}=\sum_{n=0}^{\infty}\left(\frac{\frac{1}{2}z^{n}}{z^{n}-i}+\frac{\frac{1}{2}z^{n}}{z^{n}+i}\right). $

The trouble with this idea is that we know that if the two series in this decompostion converge, then their sum converges. The converse isn't true though...

Exercise 2

If $ f $ is analytic on the unit disc and

$ \left| f(z)\right|\leq \frac{1}{1-\left| z\right|}, $

find the best estimate of

$ \left| f^{(n)}(0)\right| $

thet the Cauchy Estimates will yield.

Discussion

-The exercise seems relatively straightforward: Express the bound given by the Cauchy Estimates as a function of one real variable and then use standard calculus optimization techniques to find the minimum. Anyone have a different idea? Any slick proofs?

Exercise 3

Show that the successive derivatives of an analytic function at a point $ a $ can never satisfy

$ \left|f^{(n)}(a)\right|>n!n^{n}. $

Formulate a sharper theorem of the same kind.

Discussion

Back to MA530, Spring 2014