18.112 Functions of A Complex Variable: Mit Opencourseware

Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

MIT OpenCourseWare

http://ocw.mit.edu

18.112 Functions of a Complex Variable


Fall 2008

For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.
Lecture 3: Analytic Functions; Rational Functions

(Text 21-32)

Remarks on Lecture 3

� Formula (14) on p.32 was proved under the assumption that R(≡) = ≡. On the
other hand, if R(≡) is finite, then (12) holds with G � 0. Then we use the previous
proof on R(�j + �1 ) and we still get the representation (14).

� For theorem 1 on page 29, we have the following stronger version:

Theorem 1 (Stronger version) The smallest convex set which contains all the
zeros of P (z) also contains the zeros of P � (z).

Proof: Let �1 , · · · , �n be the zeros of P , so

P (z) = an (z − �1 ) · · · (z − �n ).

Then
P � (z) 1 1
= +···+ .
P (z) z − �1 �n
If z0 is a zero of P � (z) and z0 =≥ each �i , then this vanishes for z = z0 ; conjugating
the equation gives
z0 − � 1 z0 − � n
+···+ = 0,
|z0 − �1 | 2 |z0 − �n |2
so
z 0 = m 1 �1 + · · · + m n �n ,
where n

mi � 0 and mi = 1.
i=1

We now only need to prove the following simple result:

1
Proposition 1 Given a1 , · · · , an ≤ C, the set
n
� n

{ mi ai | mi � 0, mi = 1} (1)
i=1 i=1

is the intersection C of all convex sets containing all ai (which is called the convex
hull of a1 , · · · , an ).

n

Proof: We must show that each point ai mi in (1) is contained in each convex set
i=1
containing the ai and thus in C. We may assume it has the form
p

x= mi a i
i=1

where
mi > 0 for 1 � i � p
and
mj = 0 for j > p.
We prove x ≤ C by induction on p. Statement is clear if p = 1. Put
p−1

�= mi
i=1

and
p−1
� m1
a= ai .
i=1

By inductive assumption, a ≤ C. But
p

x= mi ai = �a + (1 − �)ai
i=1

where 0 � � � 1. So x ≤ C as stated. Q.E.D.

Solution to 4 on p.33

Suppose R(z) is rational and


|R(z)| = 1

for |z| = 1. Then


|R(ei� )| � 1 λ ≤ R.
Let S(z) be the rational functions obtained by conjugating all the coefficients in
R(z), then
R(ei� )S(e−i� ) = R(ei� )R(ei� ) = 1.
So
1
R(z)S( ) = 1 on |z| = 1.
z
Clearing denominators we see this relation
1
R(z)S( ) = 1
z
holds for all z ≤ C.
Since a polynomial has only finitely many zeroes, let
�1 , · · · , �p
be all the zeroes of R(z) which are not equal to 0 or ≡. Then
1 1
,··· ,
�1 �p
are the poles of S(z) which are not equal to 0 or ≡. So
1 1
,··· ,
¯1
� ¯p

are the poles of R(z) which are not equal to 0 or ≡ because of the definition of S.
Then � �−1
z − �1 z − �p
R(z) ···
1−� ¯1 z 1−�¯pz
has no poles or zeros except possibly 0 and ≡. Hence
z − �1 z − �p
R(z) = Cz l ···
1−�¯1z 1−�¯pz
where C is constant with |C| = 1, l is integer.
Conversely, such R has |R(z)| = 1 on |z| = 1.

You might also like