Structure theorem for Gaussian measures: Difference between revisions

Revision as of 03:44, 2 February 2023

Latest revision as of 16:22, 9 May 2025

(5 intermediate revisions by 4 users not shown)

Line 1:

{{Short description|Mathematical theorem}}

In [[mathematics]], the '''structure theorem for Gaussian measures''' shows that the [[abstract Wiener space]] construction is essentially the only way to obtain a [[Strictly positive measure|strictly positive]] [[Gaussian measure]] on a [[separable space|separable]] [[Banach space]]. It was proved in the 1970s by [[Gopinath Kallianpur|Kallianpur]]&ndash;SatoSatô&ndash;Stefan and [[Richard M. Dudley|Dudley]]&ndash;[[Jacob Feldman|Feldman]]&ndash;[[Lucien le Cam|le Cam]].<!-- I don't see those papers cited here. -->

There is the earlier result due to H. Satô (1969) <ref>[http://projecteuclid.org/DPubS/Repository/1.0/Disseminate?view=body&id=pdf_1&handle=euclid.nmj/1118797795 H. Satô, Gaussian Measure on a Banach Space and Abstract Wiener Measure], 1969.</ref> which proves that "any Gaussian measure on a separable Banach space is an [[abstract Wiener space | abstract Wiener measure]] in the sense of [[Leonard Gross | L. Gross]]". The result by Dudley et al. generalizes this result to the setting of Gaussian measures on a general [[topological vector space]].

Line 5 ⟶ 6:

==Statement of the theorem==

Let ''γ'' be a strictly positive Gaussian measure on a separable Banach space (''E'',&nbsp;||&nbsp;||). Then there exists a separable [[separable Hilbert space]] (''H'',&nbsp;&lang;&nbsp;,&nbsp;&rang;) and a map ''i''&nbsp;:&nbsp;''H''&nbsp;→&nbsp;''E'' such that ''i''&nbsp;:&nbsp;''H''&nbsp;→&nbsp;''E'' is an abstract Wiener space with ''γ''&nbsp;=&nbsp;''i''_∗(''γ''^''H''), where ''γ''^''H'' is the [[canonical form|canonical]] Gaussian [[cylinder set measure]] on ''H''.

==References==

{{Reflistreflist}}

* {{cite journal

Line 27 ⟶ 28:

{{Banach spaces}}

[[Category:ProbabilityBanach theoremsspaces]]

[[Category:Theorems in probability theory]]

[[Category:Theorems in measure theory]]