Note: a Dynkin system is also called a "d-system"[1] and the page d-system just redirects here.
Definition
\newcommand{\bigudot}{ \mathchoice{\mathop{\bigcup\mkern-15mu\cdot\mkern8mu}}{\mathop{\bigcup\mkern-13mu\cdot\mkern5mu}}{\mathop{\bigcup\mkern-13mu\cdot\mkern5mu}}{\mathop{\bigcup\mkern-13mu\cdot\mkern5mu}} }\newcommand{\udot}{\cup\mkern-12.5mu\cdot\mkern6.25mu\!}\require{AMScd}\newcommand{\d}[1][]{\mathrm{d}^{#1} }Given a set X and a family of subsets of X, which we shall denote \mathcal{D}\subseteq\mathcal{P}(X) is a Dynkin system[2] if:
- X\in\mathcal{D}
- For any D\in\mathcal{D} we have D^c\in\mathcal{D}
- For any (D_n)_{n=1}^\infty\subseteq\mathcal{D} is a sequence of pairwise disjoint sets we have \udot_{n=1}^\infty D_n\in\mathcal{D}
Given a set X and a family of subsets of X we denote \mathcal{D}\subseteq\mathcal{P}(X) is a Dynkin system[3] on X if:
- X\in\mathcal{D}
- \forall A,B\in\mathcal{D}[B\subseteq A\implies A-B\in\mathcal{D}]
- Given a sequence (A_n)_{n=1}^\infty\subseteq\mathcal{D} that is increasing[Note 1] and has \lim_{n\rightarrow\infty}(A_n)=A we have A\in\mathcal{D}
Proof of equivalence of definitions
[Expand]
Claim: Definition 1 \iff Definition 2
\newcommand{\bigudot}{ \mathchoice{\mathop{\bigcup\mkern-15mu\cdot\mkern8mu}}{\mathop{\bigcup\mkern-13mu\cdot\mkern5mu}}{\mathop{\bigcup\mkern-13mu\cdot\mkern5mu}}{\mathop{\bigcup\mkern-13mu\cdot\mkern5mu}} }\newcommand{\udot}{\cup\mkern-12.5mu\cdot\mkern6.25mu\!}\require{AMScd}\newcommand{\d}[1][]{\mathrm{d}^{#1} }
TODO: Flesh out the algebra (blue boxes)
Definition 1 \implies definition 2
- Let \mathcal{D} be a subgroup satisfying definition 1, then I claim it satisfies definition 2. Let us check the conditions.
- X\in\mathcal{D} is satisfied by definition
- For A,B\in\mathcal{D} with B\subseteq A then A-B\in\mathcal{D}
- Note that A-B=(A^c\udot B)^c (this is not true in general, it requires B\subseteq AInclude ven diagram
- As by hypothesis \mathcal{D} is closed under complements and disjoint unions, we see that (A^c\udot B)^c\in\mathcal{D} thus
- we have A-B\in\mathcal{D}
- Given (A_n)_{n=1}^\infty\subseteq\mathcal{D} being an increasing sequence of subsets, we have \lim_{n\rightarrow\infty}(A_n)=A where A:=\bigcup_{n=1}^\infty A_n (See limit of an increasing sequence of sets for more information)
- Let (A_n)_{n=1}^\infty\subseteq\mathcal{D} be given.
- Define a new sequence of sets, (B_n)_{n=1}^\infty by:
- This is a pairwise disjoint sequence of sets.
- Now by hypothesis \bigudot_{n=1}^\infty B_n\in\mathcal{D}
- Note that \bigudot_{n=1}^\infty B_n=\bigcup_{n=1}^\infty A_n
- So we have \bigcup_{n=1}^\infty A_n\in\mathcal{D} := A, thus the limit is in \mathcal{D} - as required.
This completes the first half of the proof.
The second half isn't tricky, the only bit I recommend knowing is A\udot B=(A^c-B)^c
TODO: That second half
Immediate results
[Expand]
Proof:
- As \mathcal{D} is closed under complements and X\in\mathcal{D} by definition, X^c\in\mathcal{D}
- X^c=\emptyset so \emptyset\in\mathcal{D}
This completes the proof.
See also
Notes
- Jump up ↑ Recall this means A_{n}\subseteq A_{n+1}
References
- Jump up ↑ Probability and Stochastics - Erhan Cinlar
- Jump up ↑ Measures, Integrals and Martingales - René L. Schilling
- Jump up ↑ Probability and Stochastics - Erhan Cinlar