SOLUTIONS TO HW2

Stein and Shakarchi, Chapter 1

Exercise 13. Let f be a holomorphic function on an open set , and write f (z) = u(x, y) +
iv(x, y) as usual, with u and v differentiable.
(a) Assume <(f ) = u(x, y) is constant, so that
u u
= 0 and = 0.
x y
Since f is holomorphic, it satisfies the Cauchy-Riemann equations, and so we also get
v v
= 0 and = 0,
y x
which implies that v(x, y) is constant, and hence f is constant.
(b) The proof is the same as the one in (a).
(c) Assume |f | is constant, so that |f (z)|2 = u(x, y)2 + v(x, y)2 is also constant. We
differentiate this relation with respect to x and y and use the chain rule and the Cauchy-
Riemann equations to get the following equations:
   
u u CR eqs. u v
2 u(x, y) v(x, y) = 2 u(x, y) + v(x, y) =0
x y x x
   
u u CR eqs. u v
2 u(x, y) + v(x, y) = 2 u(x, y) + v(x, y) =0
y x y y
u(x,y) v(x,y)
Multiplying the first equation by 2
and the second equation by 2
and adding, we
obtain
u
(u(x, y)2 + v(x, y)2 ) = 0.
x
If u(x, y)2 +v(x, y)2 = 0, then each of u(x, y) and v(x, y) are 0, so the claim follows. Otherwise,
u(x, y)2 + v(x, y)2 6= 0, and we see that ux
= 0. A similar argument shows that u y
= 0, so
that u(x, y) is constant, and we conclude using part (a).
Remark. Part (c) can be done more elegantly as follows. The two equations above are
equivalent to the matrix equation
   u   
u(x, y) v(x, y) x 0
u = .
v(x, y) u(x, y) y 0

The matrix on the left has determinant u(x, y)2 +v(x, y)2 . Therefore, if u(x, y)2 +v(x, y)2 = 0,
we proceed as before, and if u(x, y)2 +v(x, y)2 6= 0, then we may invert the matrix to conclude
that u
x
= u
y
= 0.
1
2 SOLUTIONS TO HW2

Exercise 14. Let {an }N N

n=1 and {bn }n=1 be two sequences of complex number, and set Bk =
Pk
n=1 bn . Note that Bn Bn1 = bn . We then have (by reindexing the sum on the third line)

N
X N
X
an b n = an (Bn Bn1 )
n=M n=M
N
X N
X
= an Bn an Bn1
n=M n=M
N
X 1 N
X
= aN BN aM BM 1 + an Bn an Bn1
n=M n=M +1
N
X 1
= aN BN aM BM 1 + (an an+1 )Bn .
n=M

apply the previous exercise with anP= rn and bn = an .

Exercise 15. Roughly, the idea is to P
k
Let {an }
n=1 be a sequence such
Pthat n=1 an converges, and set Ak = n=1 an for k 1,

A0 = 0, and A = limn An = n=1 an . Then we have, for r < 1,

N
X N
X 1
rn an = rN AN rA0 (rn+1 rn )An
n=1 n=1
N
X 1
N
= r AN + (1 r) r n An
n=1
N
X 1
= rN AN + (1 r) r n An ,
n=0

where the last line holds since A0 = 0. Since we are interested in the limit as r approaches 1
from the left, we may assume that 0 < r < 1. Under this assumption, taking the limit as N
goes to infinity in the equation above, we obtain

X
X
rn an = (1 r) rn An .
n=1 n=0
 SOLUTIONS TO HW2 3

Now let > 0 be given, and choose N0 so that |A An | 2 for n > N0 . Then we have
   
X   X 
n n
r a A = (1 r) r A A
   
 n   n 
 n=1   n=0

 
 
?
X
n
= (1 r) r (An A)
 
 n=0


N0
!
X X
(1 r) rn |An A| + rn
n=0
2 N0 +1
N0
!
N0 +1
X r
(1 r) rn |An A| +
n=0
21r
N0
X
= (1 r) rn |An A| + rN0 +1
n=0
2
N0
X
(1 r) rn |An A| +
n=0
2
where the equality (?) follows from the identity n 1
P
PN0 n n=0 r = (1 r) when |r| < 1. Now, the
expression (1 r) n=0 r |An A| is a polynomial in r with real coefficients, which takes on
the value 0 when r = 1. Therefore, by continuity of polynomials, there exists a > 0 such
that, if 1 r < holds, we have
N0
X
(1 r) rn |An A| .
n=0
2
Therefore, taking r in the range 1 < r < 1, we obtain
 
N0
X
n
 X
r an A (1 r) rn |An A| + ,
 
2

 
n=1 n=0

which shows that limr1 n

P
n=1 r an = A.

Exercise 16. (a) When n 3, we have the inequalities

1 (log(n))2 n2 ,
which implies
1 = lim sup 1 lim sup(log(n))2/n lim sup(n1/n )2 = 1,
n n n
and therefore the radius of convergence is 1.
(b) We proceed by using the Ratio Test. We have
 
 an+1 
lim   = lim n + 1 = ,
n  an  n

and therefore the radius of convergence is 0.

(c) When n 1, we have
4n 4n + 3n 2 4n ,
4 SOLUTIONS TO HW2

which implies
n2 n2 n2
.
2 4n 4n + 3n 4n
This gives
1/n
(n1/n )2 n2 (n1/n )2

1 1
= lim sup 1/n lim sup lim sup = ,
4 n 2 4 n 4n + 3n n 4 4
which implies that the radius of convergence if 4.
(d) A more precise statement of Stirlings formula says that there exist two constants c and
d such that 0 < c < d and, for all n 1, we have
cnn+1/2 en n! dnn+1/2 en .
Therefore, we have the inequalities
c3 n3n+3/2 e3n (n!)3 d3 n3n+3/2 e3n
and
c33n+1/2 n3n+1/2 e3n (3n)! d33n+1/2 n3n+1/2 e3n ,
or, equivalently,
1
d1 33n1/2 n3n1/2 e3n c1 33n1/2 n3n1/2 e3n .
(3n)!
We deduce
c3 n (n!)3 d3 n
,
d 33n 3 (3n)! c 33n 3
which gives
1/n 1/n  3 1/n
c3 n1/n (n!)3 n1/n
 
1 d 1
= lim sup 1/n lim sup lim sup 1/n
= .
27 n d 33 3 n (3n)! n c 33 3 27
Therefore the radius of convergence is 27.
(e) Note first that if or is a negative integer, then the series has only finitely many
nonzero terms, and thus converges on all of C. Assume now that , 6 {0, 1, 2, . . .}. We
again use the ratio test. Let
( + 1) ( + n 1)( + 1) ( + n 1)
an = ;
n!( + 1) ( + n 1)
we then have
an+1 ( + n)( + n)
= .
an (n + 1)( + n)
Now choose n ||; the triangle equality and reverse triangle inequality give us
 
(n ||)(n ||)  ( + n)( + n)  (n + ||)(n + ||)
 .
(n + 1)(n + ||) (n + 1)( + n)  (n + 1)(n ||)
Therefore, we get
 
(n ||)(n ||)  lim sup (n + ||)(n + ||) = 1,
 ( + n)( + n) 
1 = lim sup lim sup 
n (n + 1)(n + ||) n (n + 1)( + n)  n (n + 1)(n ||)
and the radius of convergence is 1.
 SOLUTIONS TO HW2 5

(1)n
(f) If we let an = n!(n+r)!4n
, we get
an+1 1
= ;
an (n + 1)(n + r + 1)4
thus,  
 1 
lim sup   = 0,
n (n + 1)(n + r + 1)4 
and the radius of convergence is (that is, the function converges on the entire complex
plane).

Exercise 23. A function is infinitely differentiable if it has derivatives of all orders. Let f (x)
be defined by (
0 if x 0,
f (x) = 2
e1/x if x > 0.
It is clear that if x 6= 0, then f has derivatives of all orders at x, and is continuous in a
neighborhood of x; it therefore remains to examine the situation at x = 0. Note first that we
may compute the derivatives away from 0 by using the chain rule, and we obtain
(
0 if x < 0,
f (n) (x) = 2
e1/x Pn (x) if x > 0,
PN an
where Pn (x) is a polynomial in x1 (that is, Pn (x) = n=1 xn ). By making the change of
variables t = 1/h, we obtain
PN
1/x2 an tn
lim+ e Pn (x) = lim n=1t2 .
h0 t e
tk
The expression inside the limit is a linear combination of functions of the form et2
; applying
LHopitals rule several times, we obtain
tk k!
lim t2 = lim t2 = 0,
t e t e Qk (t)

where Qk (t) is some polynomial in t. Therefore, we conclude that

2
lim+ e1/x Pn (x) = 0.
h0
We must now compute the derivative at 0. We use the definition of derivative from the left
and right:
f (h) f (0) 00
lim = lim
h0 h h0 h
= 0,
2
f (h) f (0) e1/h
lim = lim+
h0+ h h0 h
t
= lim t2
t e
? 1
= lim 2
t 2tet
= 0,
6 SOLUTIONS TO HW2

where we have used LHopitals rule in (?). Therefore, f 0 (0) = 0, f is differentiable, and has
a continuous first derivative.
Assume now by induction that f (n1) (0) = 0. We have an expression for f (n) away from 0,
and at 0 we again use the definitions:
f (n1) (h) f (n1) (0) 00
lim = lim
h0 h h0 h
= 0,
2
f (n1)
(h) f (n1)
(0) e1/h Pn1 (h)
lim = lim+
h0+ h h0 h
tPn1 (t1 )
= lim
t et2
= 0,
where the last line follows from a computation similar to the one above. Therefore, we
conclude that f (n) (0) = 0 for all n 1, and therefore f is infintely differentiable. Moreover,
f does not have a converging power series expansion at 0, for if it did, we would necessarily
have

X f (n) (0) n
f (x) = x = 0,
n=0
n!
which is absurd.

Exercise 25. (a) We may evaluate this integral directly. Let be parametrized by z(t) =
Reit , t [0, 2]. If n 6= 1, we have
Z Z 2
n
z dz = Rn eint iReit dt
0
Z 2
n+1
= iR ei(n+1)t dt
0
 2
n+1 1
= iR ei(n+1)t
i(n + 1) 0
= 0;
if n = 1, we have
Z Z 2
1
z dz = R1 eit iReit dt
0
Z 2
= i dt
0
= 2i.
(If youre still feeling shaky about these computations, break this up into real and imaginary
parts).
(b) We can use Corollary 3.3: if does not enclose the origin, then f (z) = z n has a
1
primitive defined on the interior of . For n 6= 1, then n+1 z n+1 is a primitive, defined inside
(note that this also works for negative exponents). If n = 1, then F (rei ) = log(r) + i
 SOLUTIONS TO HW2 7

is a primitive for f (z) = z 1 (provided we restrict the domain of ). For example, if lies
entirely in the right half plane, we take [, ]. Corollary 3.3 now implies
Z
z n dz = 0.

(c) Using partial fractions, we have

1 1 1 1 1
= + .
(z a)(z b) abz a baz b
Therefore, if we use part (b) and a slight generalization of part (a), we get
Z Z Z
1 1 1 1 1 2i
dz = dz + dz = .
(z a)(z b) ab za ba zb ab

Chapter 2
2
Exercise 1. Let f (z) = ez , and let = 1 2 3 , where 1 is the straight
 line 
from 0
it 1 1
to R, 2 is the arc z(t) = Re , t [0, 4 ], and 3 is the straight line from R 2 + i 2 to 0.
Note first that, since f (z) is holomorphic, Cauchys theorem gives
Z Z Z Z
z 2 z 2 z 2 2
e dz + e dz + e dz = ez dz = 0.
1 2 3

Consider first the integral over 2 . We must show rigorously that this integral goes to 0 as
R goes to . We split the path into an upper and lower part: fix a positive real number h,
and let L denote the lower part of 2 with y coordinate less than R2 h, and let U denote
the upper part of 2 with y coordinate greater than R2 h (draw a picture to see what this
looks like). We will make use of the formula of Proposition 3.1(iii) which states
Z 
 
 f (z) dz  sup(|f (z)|) length().
  z

z 2 2) 2 +y 2
We have |f (z)| = |e | = e<(z = ex , and on 2 , we have x R

2
, and therefore
2 2 2 2

R2 /2 (R/ 2h)2
|ez
| eR /2 ey on . Now, on L , the function |ez | is bounded by e e =
Rh 2+h2 z 2
e , and the length is bounded by R 4 (the full length of 2 ). On U , |e | is bounded
R2 /2 R2 /2
by e e = 1, and the length is bounded by 2 h (notice that the length of U is bounded
by the arc length of the quarter circle of radius h).
Combining everything, we obtain
Z  Z  Z 
 ez2 dz   z 2 z 2
     
   e dz  +
 
 e dz 

2 L U
Rh2+h2
Re + h.
4 2
1
If we let h = R , this gives
Z 
 ez2 dz  Re 2R+1/R +
 
 4 ,

2 2 R
which tends to 0 as R goes to infinity.
8 SOLUTIONS TO HW2

We now examine  the integral over 3 . The inverse path 3 can be parametrized by
1 1
z(t) = t 2 + i 2 , t [0, R], and therefore
Z Z
z 2 2
e dz = ez dz
3 3
Z R  
1it2 1
= e + i dt
0 2 2
  Z R 
1 1 2 2
= + i cos(t ) i sin(t ) dt .
2 2 0

To complete the computation, we let R tend to infinity, which shows

Z Z   Z 
x2 z 2 1 1 2 2
e dx = e dz = + i cos(t ) i sin(t ) dt .
0 3 2 2 0

This implies
Z  Z  
2 2 1 1 x2 1 1
cos(t ) i sin(t ) dt = i e dx = i ;
0 2 2 0 2 2 2
comparing real and imaginary parts shows
Z Z

2 2 2
cos(t ) dt = sin(t ) dt = .
0 0 4

(iz)n1
Exercise 2. Consider the function f (z) = 2i1 e z1 , and note that, since f (z) = 12
iz P
n=1 n!
,
this function is holomorphic on all of C. We let denote the contour for which is
parametrized by z(t) = eit , t [0, ], and let R denote the contour z(t) = Reit , t [0, ].
Then Cauchys theorem gives
1 eiz 1 1 eiz 1 1 eiz 1 1 eiz 1
Z Z Z Z
dz + dz + dz + dz = 0.
[R,] 2i z 2i z [,R] 2i z R 2i z
We evaluate each of these integrals separately. Consider first the integral over . Since
f (z) is holomorphic, the quantity |f (z)| is bounded by some M in some small disc of radius
0 . Taking < 0 , we obtain
iz
Z 
 1 e 1 
 M length( ) = M .
 dz
2i z
 

Note that as long as < 0 , the bound on supz (|f (z)|) is independent of . Therefore,
letting tend to 0, we obtain
1 eiz 1
Z
dz 0.
2i z
We now consider the integral over R , again by splitting R into a lower and upper part.
We again fix a positive real number h, and let L denote the part of R with y coordinate
less than h, and let U denote the part of R with y coordinate greater than h. We have
|eiz | = e<(iz) = ey , and therefore on L the maximum of this function is 1, and on U the
maximum is eh . The length of L is bounded by h (since the length of the two small arc
 SOLUTIONS TO HW2 9

is certainly bounded above by the length of half-circle of radius h), and the length of U is
bounded by R. Therefore, we obtain
1 eiz 1 eiz 1 eiz
Z  Z Z 
   
 dz  = 
  dz + dz 
R 2i z 2i z U 2i z

Z L iz
  Z iz

 1 e   1 e 
 dz  +  dz 
2i z
L U 2i z
h
1 1e
h + R
2R 2 R
h eh
= + .
2R 2

Therefore, taking h = R, we get

iz R
Z 
 1 e  e
 dz  + ,

R 2i z 2 R 2
which tends to 0 as R goes to infinity. On the other hand, using the parametrization z(t) =
Reit , we get
1
Z Z
1
dz = dt
R 2iz 2 0

= .
2
Therefore, as R goes to infinity, we obtain
1 eiz 1
Z

dz .
R 2i z 2
Upon combining everything and letting R tend to infinity and tend to 0, we get
Z
1 eix 1
dx = ;
2i x 2
taking the real part of both sides shows
Z
1 sin(x)
Z
sin(x)
dx = dx = .
0 x 2 x 2
Remark. Let P (z) and Q(z) be two polynomials such that deg(P (z)) + 1 deg(Q(z)). Then
the above proof actually shows that we have
Z
P (z) iz
e dz 0
R Q(z)

as R tends to infinity.

Exercise
3. Let a, b be two positive real numbers (we can always assume b is positive), let
A = a2 + b2 , and let 0 2 be the unique angle for which cos() = Aa . Note that this
implies sin() = Ab . Let = 1 2 3 , where 1 is the straight line from 0 to R, 2 is the
10 SOLUTIONS TO HW2

curve parametrized by z(t) = Reit , t [0, ], and 3 is the straight line from Rei to 0. By
Cauchys theorem, we have
Z Z Z Z
Az Az Az
e dz + e dz + e dz = eAz dz = 0.
1 2 3

We first evaluate the integral along 2 . We have

z=Rei
eAz
Z 
Az
e dz =
2 A z=R
AR ARei
e e
=
A
eAR eR(a+ib)
= ;
A
therefore
 eAR + eRa
Z 
Az

 e dz  ,

2
 A
which tends to 0 as R tends to infinity.
Thus, letting R tend to infinity, we obtain
Z Z
Az
e dz = (cos() + i sin())eA(cos()+i sin())t dt
3 0
Z  
a b
= +i eatibt dt
A A
Z0
= eAt dt
0 At 
e
=
A t=0
1
= .
A
a b
Equating real and imaginary parts (and using the relations cos() = A
and sin() = A
),
we get the pair of equations

a at b at
Z Z
1
e cos(bt) dt + e sin(bt) dt =
A 0 A 0 A
b at a at
Z Z
e cos(bt) dt e sin(bt) dt = 0
A 0 A 0

Writing this in matrix form, we get

a b
 R at  1
A A
e
R0 at cos(bt) dt
b = A .
A
Aa 0
e sin(bt) dt 0
 SOLUTIONS TO HW2 11

Inverting this, we get

Z
a a
eat cos(bt) dt =2
= 2 ,
A a + b2
Z0
b b
eat sin(bt) dt = = .
0 A2 a2 + b 2