Namespaces
Variants
Actions

Difference between revisions of "User:Boris Tsirelson/sandbox1"

From Encyclopedia of Mathematics
Jump to: navigation, search
Line 16: Line 16:
 
That  is, up to almost isomorphism we have "the" atomless standard probability space. Its "incarnations" include the spaces $\R^n$ with atomless probability distributions (be they [[Continuous distribution|absolutely continuous]] or [[Singular distribution|singular]]), as well as the set of all continuous functions $[0,\infty)\to\R$ with the [[Wiener measure]]. That is instructive: topological notions such as dimension, connectedness, compactness etc. do not apply to probability spaces.
 
That  is, up to almost isomorphism we have "the" atomless standard probability space. Its "incarnations" include the spaces $\R^n$ with atomless probability distributions (be they [[Continuous distribution|absolutely continuous]] or [[Singular distribution|singular]]), as well as the set of all continuous functions $[0,\infty)\to\R$ with the [[Wiener measure]]. That is instructive: topological notions such as dimension, connectedness, compactness etc. do not apply to probability spaces.
  
====Measure preserving injections====
+
====Measure preserving maps====
 +
 
 +
The inverse to a bijective measure preserving map is measure preserving provided that it is measurable; in this (not general!) case the given map is a strict isomorphism.
  
 
Here is another important fact in two equivalent forms.
 
Here is another important fact in two equivalent forms.
  
'''Theorem 2a.''' If a bijective map between standard probability spaces is measure preserving then the inverse map is also measure preserving.
+
'''Theorem 2a.''' Every bijective measure preserving map between standard probability spaces is a strict isomorphism.
  
 
'''Theorem 2b.''' If $(\Om,\F,P)$ is a standard probability space and $\F_1\subset\F$ a sub-σ-field such that $(\Om,\F_1,P|_{\F_1})$ is standard then $\F_1=\F$.  
 
'''Theorem 2b.''' If $(\Om,\F,P)$ is a standard probability space and $\F_1\subset\F$ a sub-σ-field such that $(\Om,\F_1,P|_{\F_1})$ is standard then $\F_1=\F$.  
  
Recall a topological fact similar to Theorem 2: if a bijective map  between compact Hausdorff topological spaces is continuous then the  inverse map is also continuous. Moreover, if a Hausdorff topology is  weaker than a compact topology then these two topologies are equal,  which has the following probability-space counterpart stronger than Theorem 2  (in two equivalent forms).  
+
Recall a topological fact similar to Theorem 2: if a bijective map  between compact Hausdorff topological spaces is continuous then it is a homeomorphism. Moreover, if a Hausdorff topology is  weaker than a compact topology then these two topologies are equal,  which has the following probability-space counterpart stronger than Theorem 2  (in two equivalent forms).  
  
 
'''Theorem 3a.''' If a bijective map from a standard probability space to a  countably separated probability space is measure preserving then the inverse map  is also measure preserving.  
 
'''Theorem 3a.''' If a bijective map from a standard probability space to a  countably separated probability space is measure preserving then the inverse map  is also measure preserving.  
 
(In general, the inverse to a bijective measure preserving map is measure preserving provided that it is measurable.)
 
  
 
'''Theorem 3b.''' If $(\Om,\F,P)$ is a standard probability space and $\F_1\subset\F$ is a countably separated sub-σ-field then $(\Om,\F,P)$ is the completion of $(\Om,\F_1,P|_{\F_1})$.
 
'''Theorem 3b.''' If $(\Om,\F,P)$ is a standard probability space and $\F_1\subset\F$ is a countably separated sub-σ-field then $(\Om,\F,P)$ is the completion of $(\Om,\F_1,P|_{\F_1})$.

Revision as of 09:31, 1 March 2012

$\newcommand{\Om}{\Omega} \newcommand{\F}{\mathcal F} \newcommand{\B}{\mathcal B} \newcommand{\M}{\mathcal M} $ A probability space is called standard if it satisfies the following equivalent conditions:

The isomorphism theorem

Every standard probability space consists of an atomic (discrete) part and an atomless (continuous) part (each part may be empty). The discrete part is finite or countable; here, all subsets are measurable, and the probability of each subset is the sum of probabilities of its elements.

Theorem 1. All atomless standard probability spaces are mutually almost isomorphic.

That is, up to almost isomorphism we have "the" atomless standard probability space. Its "incarnations" include the spaces $\R^n$ with atomless probability distributions (be they absolutely continuous or singular), as well as the set of all continuous functions $[0,\infty)\to\R$ with the Wiener measure. That is instructive: topological notions such as dimension, connectedness, compactness etc. do not apply to probability spaces.

Measure preserving maps

The inverse to a bijective measure preserving map is measure preserving provided that it is measurable; in this (not general!) case the given map is a strict isomorphism.

Here is another important fact in two equivalent forms.

Theorem 2a. Every bijective measure preserving map between standard probability spaces is a strict isomorphism.

Theorem 2b. If $(\Om,\F,P)$ is a standard probability space and $\F_1\subset\F$ a sub-σ-field such that $(\Om,\F_1,P|_{\F_1})$ is standard then $\F_1=\F$.

Recall a topological fact similar to Theorem 2: if a bijective map between compact Hausdorff topological spaces is continuous then it is a homeomorphism. Moreover, if a Hausdorff topology is weaker than a compact topology then these two topologies are equal, which has the following probability-space counterpart stronger than Theorem 2 (in two equivalent forms).

Theorem 3a. If a bijective map from a standard probability space to a countably separated probability space is measure preserving then the inverse map is also measure preserving.

Theorem 3b. If $(\Om,\F,P)$ is a standard probability space and $\F_1\subset\F$ is a countably separated sub-σ-field then $(\Om,\F,P)$ is the completion of $(\Om,\F_1,P|_{\F_1})$.

A continuous image of a compact topological space is always a compact set, which also has a probability-space counterpart.

Theorem 4. Let $(\Om,\F,P)$ be a standard probability space, $(\Om_1,\F_1,P_1)$ a countably separated complete probability space, and $f:\Om\to\Om_1$ a measure preserving map. Then $(\Om_1,\F_1,P_1)$ is also standard, and the image $f(\Om)$belongs to $\F_1$. (See [R, Th. 3-2].)



Non-example. The set $[0,1]^\R$ of all functions $\R\to[0,1]$ with the product of Lebesgue measures is a nonstandard probability space.

Definition 1a. A probability space $(\Om,\F,P)$ is standard if it is complete and there exist a subset $\Om_1\subset\Om$ and a σ-field (in other words, σ-algebra) $\B$ on $\Om_1$ such that $(\Om_1,\B)$ is a standard Borel space and every set of $\F$ is almost equal to a set of $\B$. (See [I, Sect. 2.4].) (Clearly, $\Om_1$ must be of full measure.)

Definition 1b (equivalent). A probability space $(\Om,\F,P)$ is standard if it is complete, perfect and countably separated mod 0 in the following sense: some subset of full measure, treated as a subspace of the measurable space $(\Om,\F)$, is a countably separated measurable space.

(See [I, Sect. 3.1] for a proof of equivalence of these definitions.)

On terminology

Also "Lebesgue-Rokhlin space" and "Lebesgue space".

In [M, Sect. 6] universally measurable spaces are called metrically standard Borel spaces.

In [K, Sect. 21.D] universally measurable subsets of a standard (rather than arbitrary) measurable space are defined.

In [N, Sect. 1.1] an absolute measurable space is defined as a separable metrizable topological space such that every its homeomorphic image in every such space (with the Borel σ-algebra) is a universally measurable subset. The corresponding measurable space (with the Borel σ-algebra) is also called an absolute measurable space in [N, Sect. B.2].

References

[I] Kiyosi Itô, "Introduction to probability theory", Cambridge (1984).   MR0777504   Zbl 0545.60001
[B] V.I. Bogachev, "Measure theory", Springer-Verlag (2007).   MR2267655  Zbl 1120.28001
[C] Donald L. Cohn, "Measure theory", Birkhäuser (1993).   MR1454121   Zbl 0860.28001
[D] Richard M. Dudley, "Real analysis and probability", Wadsworth&Brooks/Cole (1989).   MR0982264   Zbl 0686.60001
[M] George W. Mackey, "Borel structure in groups and their duals", Trans. Amer. Math. Soc. 85 (1957), 134–165.   MR0089999   Zbl 0082.11201
[K] Alexander S. Kechris, "Classical descriptive set theory", Springer-Verlag (1995).   MR1321597   Zbl 0819.04002
[N] Togo Nishiura, "Absolute measurable spaces", Cambridge (2008).   MR2426721   Zbl 1151.54001
How to Cite This Entry:
Boris Tsirelson/sandbox1. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Boris_Tsirelson/sandbox1&oldid=21377