Math 220, Practice Problems For The Final Exam.: N N N N
Math 220, Practice Problems For The Final Exam.: N N N N
To review, we list below the Main Topics covered in this class (this is not a compre-
hensive list):
(1) Holomorphic functions. Harmonic functions.
(2) Existence of local/global primitives. Logarithm. Winding numbers.
(3) Cauchy’s integral formula. Cauchy’s estimates.
(4) Taylor and Laurent series.
(5) Zeroes of holomorphic functions, open mapping theorem, maximum modulus
principle, Liouville’s theorem.
(6) Types of singularities. Removable singularities theorem. Meromorphic functions.
Residues. Cassorati-Weierstraβ.
(7) Residue theorem. Residues at infinity. Applications to real analysis.
(8) The argument principle. Applications to elliptic functions. Rouché’s theorem.
(9) Sequences of holomorphic functions. Hurwitz’s theorem. Weierstraβ convergence
theorem.
(10) Infinite products of holomorphic functions. Factorization of sine. Weierstraβ
factorization.
1. Let U be open and connected, and let f, g be holomorphic functions such that
f (z)g(z) = 0. Show that either f or g is identically zero on U .
2.
(i) Let x ∈ C. Show that the Laurent expansion
∞
(−1)n
1 1 X
n
exp x z− = J0 (x) + Jn (x) z +
2 z zn
n=1
holds for 0 < |z| < ∞ for some coefficients Jn (x) that depend on x.
Remark: These coefficients Jn are called the Bessel functions of the first kind,
and appear for instance in the study of the wave equation.
(ii) Using the expansion of the exponential, show that Jn are entire and
∞
X (−1)k x n+2k
Jn (x) = .
k!(k + n)! 2
k=0
(iii) Show that y = Jn (x) is a solution to the Bessel differential equation
x2 y 00 + xy 0 + (x2 − n2 )y = 0.
3. Assume f and g are meromorphic functions on C such that
|f (z) − g(z)| < |g(z)|
for all z ∈ C which are not poles for f or g. Show that f = cg for some constant c. One
possible approach is via removable singularity theorem.
1
2
5. Show that there exists a holomorphic function f on the unit disc ∆(0, 1) that
cannot be extended to a holomorphic function in the neighborhood of any point on the
boundary ∂∆.
(ii) Show that σ is an entire function with zeroes only at the points of Λ.
8. Exhibit an entire function which vanishes only at the positive integers n with order
n, and is non-zero everywhere else.
3
9. Assume that f and g are entire functions. Show that there exist entire functions
h, F and G such that
f (z) = h(z)F (z), g(z) = h(z)G(z)
with F, G having no common zeroes. This is an application of the Weierstraβ problem.
10. Possibly using the product expansion of the sine function, prove that
∞
eaz − ebz (a − b)2 z 2
Y
(a+b)z/2
= ze 1+ .
a−b 4n2 π 2
n=1
11. Prove that the polynomial az 5 + z + 1 has at least one root inside the disc |z| ≤ 2.
1 1
Hint: there are two cases a ≤ 32 and a > 32 . One can be treated using Rouché. The
other case follows by looking at the product of roots.
f (n) (z)
12. Let f (z) be a holomorphic function in the disk |z| < 2. Show that ∞
P
n=1 n!
converges locally uniformly over |z| < 1. You may wish to recall Cauchy’s estimates.
13. Assume that f : ∆(0, 2) → C is a holomorphic function such that |f (z)| ≤ M for
|z| = 1. Using Cauchy’s formula, show that for all w1 , w2 ∈ ∆(0, 12 ) we have
|f (w1 ) − f (w2 )| ≤ 4M |w1 − w2 |.
14.
(i) Let A = {z : |z| ≤ R}, and let f be a holomorphic function in a neighborhood of
A. Show that for all > 0, there exists a polynomial p such that
sup |p(z) − f (z)| < .
z∈A
(ii) Assume that A = {z : r ≤ |z| ≤ R} for R > r > 0. Show that there exists > 0
such that for all polynomials p we have
z
e
sup p(z) − > .
z∈A z
ez
That is, show that z cannot be approximated by polynomials uniformly on A.
This is an application of integration.
15.
(i) Show that C(z) = z−i z+i takes the upper half plane bijectively onto the unit disc;
in particular, if Im z > 0 then |C(z)| < 1.
(ii) Conclude from (i) that there are no entire functions with f : C → C such that
Im f (z) > 0 for all z ∈ C.
Remark: This was part of an older problem set, but solved in a different way
then.
Hint: You may wish to consider the function g(z) = f (z)/z for z 6= 0, g(0) =
f 0 (0) and use the maximum modulus principle over any disc ∆(0, r). Make r → 1.
(ii) Show that if f has a fixed point α 6= 0 inside ∆(0, 1) then f is the identity.