-
Notifications
You must be signed in to change notification settings - Fork 1
/
122A Homework 3 - Analytic, Complex Differentiable.tex
228 lines (167 loc) · 8.67 KB
/
122A Homework 3 - Analytic, Complex Differentiable.tex
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
\documentclass[11pt]{article}
\usepackage[margin=1.25in]{geometry}
\usepackage{graphicx,tikz}
\usepackage{amsmath,amsthm}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{boondox-cal}
\title{ }
\newtheorem*{thm}{Theorem}
\begin{document}
%\maketitle
%\date
\begin{center} % centers
\Large{Homework 3} % Large makes the font larger, put title inside { }
\end{center}
\begin{center}
Vincent La \\
Math 122A \\
July 15, 2017
\end{center}
\section{Monday}
\begin{enumerate}
\item[Extra 5.] Write down in terms of $M$-neighborhoods of $\infty$ what it means that $\lim_{z \rightarrow \infty} f(z) = \infty$.
\paragraph{Answer} When we project the complex numbers on a sphere, we can see that as the magnitude of the number increases more and more they seem to converge to the north pole of the sphere.
\item[Extra 6.] Prove that $\lim_{z \rightarrow 0}{\frac{1}{z}} = \infty$ and that $\lim_{z \rightarrow 0} \frac{1}{z} = \infty$ and that $\lim_{z \rightarrow \infty} \frac{1}{z} = 0$.
\begin{enumerate}
\item Prove that $\lim_{z \rightarrow 0}{\frac{1}{z}} = \infty$
\begin{proof}
Let $M > 0$, $z \in \mathbb{C}$ be arbitrary. Then pick $\delta = \frac{1}{M}$.
As a result, whenever $|z - 0| = |z| < \delta$, we get that
\[|z| < \frac{1}{M}\]
Now, multiply both sides by $M$, and we get
\[|z|M < 1\]
Divide both sides by $|z|$, and we get
\[M < \frac{1}{|z|} = f(z)\]
as we set out to prove.
\end{proof}
\item Prove that $\lim_{z \rightarrow \infty} \frac{1}{z} = 0$
\begin{proof}
We want to show that for any $\epsilon > 0$, there is a $\delta > 0$ so that if $|z| > \delta$ then $|f(z) - 0| < \epsilon$. Let $\delta = \frac{1}{\epsilon} >0$.
\bigskip
If $|z| > \delta$, then $|z| > \frac{1}{\epsilon}$. If we take the reciprocal of both sides, we get
\[f(z) = \frac{1}{|z|} < \epsilon \]
as we set out to prove.
\end{proof}
\end{enumerate}
\item[Extra 7.] A complex values function $f: D \rightarrow \mathbb{C}$ is called bounded if there exists $M > 0$ such that $|f(z)| < M$ for all $z \in D$. Prove that $\{z_n\} \subseteq D$ such that $f(z_n) \rightarrow \infty$.
\end{enumerate}
\section{Tuesday}
\begin{enumerate}
\item[2.3] By comparing coefficients or by use of the Cauchy-Riemann equations, determine which of the following polynomials are analytic.
\begin{enumerate}
\item $P(x + iy) = x^3 - 3xy^2 - x + i(3x^2y - y^3 - y)$
\paragraph{Analytic}
Here,
\[u(x, y) = x^3 - 3xy^2 - x\] and
\[v(x, y) = 3x^2y - y^3 - y\]. Therefore,
\[\begin{aligned}
u_x &= 3x^2 - 3y^2 - 1 \\
v_y &= 3x^2 - 3y^2 - 1 \\
u_y &= -6xy \\
-v_x &= -6xy
\end{aligned}
\]
Because $u_x = v_y$ and $u_y = -v-x$, $P(x + iy)$ is analytic.
\item $P(x + iy) = x^2 + iy^2$
\paragraph{Not Analytic}
Here, $u(x, y) = x^2$ and $v(x, y) = y^2$. Therefore,
\[\begin{aligned}
u_x &= 2x \\
v_y &= 2y \\
\end{aligned}\]
implying that $u_x \neq v_y$ or that $P(x + iy)$ is not analytic.
\item $P(x + iy) = 2xy + i(y^2 - x^2)$
\paragraph{Analytic}
Here, $u(x, y) = 2xy$ and $v(x, y) = y^2 - x^2$. Therefore,
\[\begin{aligned}
u_x &= 2y \\
v_y &= 2y \\
u_y &= 2x \\
-v_x = -(-2x) = 2x \\
\end{aligned}\]
Because $u_x = v_y$ and $u_y = -v_x$, $P(x + iy)$ is analytic.
\end{enumerate}
\item[2.4] Show that no nonconstant analytic polynomial can take imaginary values only.
\begin{proof}
Let us prove the contrapositive, that if $f$ takes only imaginary values then $f$ is either constant or not analytic.
\bigskip
If $f$ only takes imaginary values then it has no real component, i.e. $u(x, y) = 0$. Thus, $u_x = 0$ and $u_y = 0$. If $f$ is analytic, then necessarily $v_x = v_y = 0$. Because, only a constant function has these derivatives $f$ is constant.
\bigskip
Now, if we relax the requirement that $f$ satisfy the Cauchy-Riemann equations, then it is free to be non-constant. However, it will not be analytic, as we set out to prove.
\end{proof}
\end{enumerate}
\section{Wednesday and Thursday}
\begin{enumerate}
\item[3.2] Show $f(z) = x^2 + iy^2$ is differentiable at all points on the line $x = y$.
\begin{proof}
Let $z \in \mathbb{C}$ be arbitrary as long as $x = y$. Then, we want to show that $f$ is differentiable, or in other words $\lim_{h \rightarrow 0} \frac{f(z+h) - f(z)}{h}$ exists.
Let us write $z = x_1 + iy_1$ and $h \in \mathbb{C}$ as $h = x_2 + iy_2$. Then, our limit becomes
\[\begin{aligned}
& \lim_{x_2 + iy_2 \rightarrow 0} \frac{
f(x_1 + iy_1 + x_2 + iy_2) -
f(x_1 + iy_1)}{
x_2 + iy_2} \\
& \lim_{x_2 + iy_2 \rightarrow 0} \frac{
f[x_1 + x_2 + i(y_1 + y_2)] -
f(x_1 + iy_1)}{
x_2 + iy_2} \\
& \lim_{x_2 + ix_2 \rightarrow 0} \frac{
f[x_1 + x_2 + i(x_1 + x_2)] -
f(x_1 + ix_1)}{
x_2 + ix_2} \\
& \lim_{x_2 + ix_2 \rightarrow 0} \frac{
(x_1 + x_2)^2 + i(x_1 + x_2)^2 -
(x_1^2 + ix_1^2)}{
x_2 + ix_2} \\
& \lim_{x_2 + ix_2 \rightarrow 0} \frac{
x_1^2 + 2x_1x_2 + x_2^2 + ix_1^2 + 2ix_1x_2 + ix_2^2 -
(x_1^2 + ix_1^2)}{
x_2 + ix_2} \\
& \lim_{x_2 + ix_2 \rightarrow 0} \frac{
2x_1x_2 + x_2^2 + 2ix_1x_2 + ix_2^2}{
x_2 + ix_2} \\
& \lim_{x_2 + ix_2 \rightarrow 0} \frac{
2x_1x_2(1 + i) + x_2^2(1 + i)}{
x_2(1 + i)} \\
& \lim_{x_2 + ix_2 \rightarrow 0} \frac{
2x_1x_2 + x_2^2}{
x_2} \\
& \lim_{x_2 + ix_2 \rightarrow 0} 2x_1 + x_2 = 2x_1 + x_2 \\
\end{aligned}\]
Therefore, our limit exists so $f$ is differentiable.
\paragraph{Nowhere Analytic} Show that $f$ is nowhere analytic.
Using the Cauchy-Riemann Equations, we see that
\[u_x = 2y, v_y = 2y, u_y =0, -v_x = 0\]
Therefore, $f$ is analytic only on the line $x = y$. However, for $f$ to be analytic for any $z$, it must be analytic in a neighborhood of $z$. However, for any $z$ where $f$ is analytic, its neighborhood contains an infinite number of points where $x \neq y$. Therefore, $f$ is nowhere analytic.
\end{proof}
\item[3.5] Suppose $f$ is analytic in a region and $f' \equiv 0$ there. Show that $f$ is constant.
\begin{proof}
Let $f$ be any function analytic in a region where $f' \ equiv 0$ in that region. Because it is analytic, it satisfies the Cauchy-Riemann Equations, implying $u_x = v_y$ and $u_y = -v_x$. Because $f' \equiv 0$, it must mean that its partial derivatives either cancel each other out or are equal to 0. However, because $f$ is analytic, the conditions imposed by the Cauchy-Riemann Equations imply $u_x = v_y = 0$ and $u_y = -v_x =0$. Only a constant function has these derivatives, so $f$ is constant.
\end{proof}
\item[3.9] Show that there are no analytic functions $f = u + iv$ with $u(x, y) = x^2 + y^2$.
\begin{proof}
Let $f$ be an arbitrary function with $u(x, y) = x^2 + y^2$, and assume for a contradiction that it is analytic.
\bigskip
Because $f$ is analytic, it satisfies the Cauchy-Riemann Equations. Because $u_x = 2x$ and $u_y = 2y$, this implies $v_y = 2x$ and $-v_x = 2y$. Now, notice that $v_y = 2x$ implies
\[\int v_y dy = \int 2x dy\]
\[v = 2xy + c(y) \]
On the other hand, $v_x = -2y$ implies
\[\int v_x dx = \int -2y dx \]
\[v = -2xy + c(x)\]
However, $v = 2xy + c(y)$ and $v = -2xy + c(x)$ cannot be true at the same time. Therefore, there are no analytic functions $f = u + iv$ with $u(x, y) = x^2 + y^2$.
\end{proof}
\item[3.10] Suppose $f$ is an entire function of the form
\[f(x, y) = u(x) + iv(y)\]
Show that $f$ is a linear polynomial.
\begin{proof}
Let $f$ be any entire function of the form $u(x) + iv(y)$. If $f$ is entire, it is analytic anywhere, so it constantly satisfies the Cauchy Riemann equations. Because $u_y = -v_x = 0$ for all $x, y$ this imposes the constraint that $u_x = v_y$ for all $x, y$.
\bigskip
Suppose for a contradiction that $u(x), v(y)$ are not linear polynomials, i.e. they are polynomials with degree $> 1$. However, if we take partial derivatives $u_x$ and $v_y$, then they will necessarily have some term involving $x$ and $y$ respectively. Now, take for example--$x = 0, y = 1$, then their partial derivatives evaluate to $u_x= 0, v_y= 1$. This contradicts the assumption $f$ is entire, so this cannot be the case.
\bigskip
Now, we are left with the possibility that $u(x), v(y)$ are both linear polynomials. Write $u(x) = a + bx$ and $v(y) = c + dy$. Taking partial derivatives, we get $u_x = b$ and $v_y = d$. Thus, the Cauchy-Riemann Equations are satisfied if $x$ and $y$ have the same coefficients.
\bigskip
In conclusion, we have shown that any entire function of the form $u(x) + iv(y)$ has $u(x)$ and $v(y)$ as linear polynomials sharing the same coefficient on the first degree term.
\end{proof}
\end{enumerate}
\end{document}