aboutsummaryrefslogtreecommitdiff
path: root/analysis_3.tex
blob: 7bc59a14bdf62f8522f2bd5c3e837aaf0cd61d2b (plain)
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
\section*{Nützliches aus der Mengenlehre}

\subsection*{De Morgansche Regeln}

Sei $\mathcal{B}$ ein Mengensystem.

$$\left(\bigcup_{B\in \mathcal{B}} B \right)^c = \bigcap_{B\in \mathcal{B}} B^c \hspace*{8mm} \left(\bigcap_{B\in \mathcal{B}} \right)^c = \bigcup_{B\in \mathcal{B}} B^c$$

\subsection*{Mengen-Ring}

Ein Mengensystem $\mathcal{A}$ ist ein Ring gdw. $\forall A, B \in \mathcal{A}$:

\begin{enumerate}[label=(\alph*)]
	\item $\emptyset \in \mathcal{A}$
	\item $B\setminus A \in \mathcal{A}$
	\item $A \cup B \in \mathcal{A}$
\end{enumerate}

\section*{$\sigma$-Algebren}

Ein Mengensystem $\mathcal{A} \subseteq \mathcal{P}(X)$ ist $\sigma$-Algebra auf der nichtleeren Menge $X$ gdw.:

\begin{enumerate}[label=(\alph*)]
	\item $X \in \mathcal{A}$
	\item $A \in \mathcal{A} \Rightarrow A^c := X\setminus A \in \mathcal{A}$
	\item $\forall j \in \mathbb{N} : A_j \in \mathcal{A} \Rightarrow \bigcup_{j\in \mathbb{N}} A_j \in \mathcal{A}$
\end{enumerate}

\subsection*{Eigenschaften von $\sigma$-Algebren}

Seien $\mathcal{A}$ eine $\sigma$-Algebra auf $X$, $n \in \mathbb{N}$, $\forall j \in \mathbb{N} : A_j \in \mathcal{A}$, dann ist $\mathcal{A}$ nach den folgenden Eigenschaften abgeschlossen unter abzählbaren Mengenoperationen:

\begin{enumerate}[label=(\alph*)]
	\item $\emptyset = X^c \in \mathcal{A}$
	\item $A_1 \bigcup \cdots \bigcup A_n \in \mathcal{A}$
	\item $A_1 \bigcap \cdots \bigcap A_n \in \mathcal{A}$
	\item $\bigcap_{j\in \mathbb{N}} A_j \in \mathcal{A}$
	\item $A_1 \setminus A_2 := A_1 \bigcap A_2^c \in \mathcal{A}$
\end{enumerate}

\subsection*{Erzeugte $\sigma$-Algebren}

Die durch das nichtleere Mengensystem $\mathcal{E} \subseteq \mathcal{P}(X)$ auf $X$ erzeugte $\sigma$-Algebra ist wie folgt definiert:

\vspace*{-4mm}
$$\sigma(\mathcal{E}) := \bigcap\{ \mathcal{A} \subseteq \mathcal{P}(X) | \mathcal{A} \text{ ist } \sigma \text{-Algebra}, \mathcal{E} \subseteq \mathcal{A} \}$$

Der Erzeuger $\mathcal{E}$ ist hierbei allg. nicht eindeutig.

\subsubsection*{Eigenschaften erzeugter $\sigma$-Algebren}

Sei $\emptyset \neq \mathcal{E} \subseteq \mathcal{P}(X)$, dann gilt:

\begin{enumerate}[label=(\alph*)]
	\item $\mathcal{A}$ ist $\sigma$-Algebra $\land$ $\mathcal{E} \subseteq \mathcal{A} \Rightarrow \mathcal{E} \subseteq \sigma(\mathcal{E}) \subseteq \mathcal{A}$
	\item $\sigma(\mathcal{E})$ ist kleinste $\mathcal{E}$ enthaltende $\sigma$-Algebra.
	\item $\mathcal{E}$ ist $\sigma$-Algebra $\Rightarrow \mathcal{E} = \sigma(\mathcal{E})$
	\item $\mathcal{E} \subseteq \mathcal{E}' \subseteq \mathcal{P}(X) \Rightarrow \sigma(\mathcal{E}) \subseteq \sigma(\mathcal{E}')$
\end{enumerate}

\subsection*{Borelsche $\sigma$-Algebra}

Sei $X$ ein metrischer Raum und $\mathcal{O}(X)$ das System der in $X$ offenen Mengen, dann ist $\mathcal{B}(X) := \sigma(\mathcal{O}(X))$ die Borelsche $\sigma$-Algebra auf $X$.

Im Speziellen wird $\mathcal{B}_m := \mathcal{B}(\mathbb{R}^m)$ gesetzt.

$\mathcal{B}_m$ enthält insb. alle offenen und abgeschlossenen Mengen in $\mathbb{R}^m$ sowie deren abzählbaren Vereinigungen und Durchschnitte.

\subsubsection*{Charakterisierung}

\vspace*{-4mm}
\begin{align*}
	\mathcal{B}_m &= \sigma(\{(a, b) | a, b \in \mathbb{Q}^m, a \leq b\}) \\
	              &= \sigma(\{(a, b] | a, b \in \mathbb{Q}^m, a \leq b\})
\end{align*}

\section*{Maße auf $\sigma$-Algebren}

Sei $\mathcal{A}$ eine $\sigma$-Algebra auf $X$.

$\mu : \mathcal{A} \rightarrow [0, \infty]$ ist positives Maß auf $\mathcal{A}$ gdw.:

\begin{enumerate}[label=(\alph*)]
	\item $\mu(\emptyset) = 0$
	\item $\forall \text{ disjunkte } \{A_j | j \in \mathbb{N}\} \subseteq \mathcal{A} :\\ \hspace*{4mm} \mu(\dot\bigcup_{j\in \mathbb{N}} A_j) = \sum_{j\in \mathbb{N}} \mu(A_j)$
\end{enumerate}

\subsection*{Maßraum}

Ein Tripel $(X, \mathcal{A}, \mu)$ ist Maßraum. Ein endlicher Maßraum erfüllt zusätzlich $\mu(X) < \infty$.

Ein Wahrscheinlichkeitsmaß erfüllt $\mu(X) = 1$.

\subsection*{Punkt- / Diracmaß}

Für fest gewählte $\mathcal{A} = \mathcal{P}(X)$, $x \in X$ ist ein Wahrscheinlichkeitsmaß für $A \subseteq X$ definiert:

$$\delta_x(A) := \begin{cases}
	1 & x \in A \\
	0 & x \notin A
\end{cases}$$

Dieses wird Punkt- / Diracmaß auf $\mathcal{A}$ genannt.

\subsection*{Zählmaß}

Sei $\mathcal{A} = \mathcal{P}(\mathbb{N})$ und $\forall j \in \mathbb{N} : p_j \in [0, \infty]$ fest gewählt.

$\mu(A) := \sum_{j\in A} p_j$ für $A \subseteq \mathbb{N}$ ist Maß auf $\mathcal{P}(\mathbb{N})$.

Gilt zusätzlich $\forall j \in \mathbb{N} : p_j = 1$ so heißt $\mu$ Zählmaß.

\subsection*{Eigenschaften von Maßen}

Sei $(X, \mathcal{A}, \mu)$ Maßraum und $A, B, A_j \in \mathcal{A}$ für $j \in \mathbb{N}$.

\begin{description}[leftmargin=!,labelwidth=26mm]
	\item[Monotonie] $A \subseteq B \Rightarrow \mu(A) \leq \mu(B)$
	\item[$\sigma$-Subadditivität] $\mu(\dot\bigcup_{j\in \mathbb{N}} A_j) \leq \sum_{j\in \mathbb{N}} \mu(A_j)$
	\item[Stetigkeit (unten)] $A_j \uparrow \Rightarrow \displaystyle\lim_{j\to \infty} \mu(A_j) = \mu(\bigcup_{j\in \mathbb{N}} A_j)$
	\item[Stetigkeit (oben)] $A_j \downarrow \land \hspace*{1mm} \mu(A_1) < \infty \\ \hspace*{4mm}\Rightarrow \displaystyle\lim_{j\to \infty} \mu(A_j) = \mu(\bigcap_{j\in \mathbb{N}} A_j)$
\end{description}

Für $\mu(A) < \infty$ folgt $\mu(B\setminus A) = \mu(B) - \mu(A)$.

Für endliche Maße gilt insb. $\mu(A^c) = \mu(X) - \mu(A)$.