Théorème de désintégration

Théorème de désintégration En mathématiques, the disintegration theorem is a result in measure theory and probability theory. It rigorously defines the idea of a non-trivial "restriction" of a measure to a measure zero subset of the measure space in question. It is related to the existence of conditional probability measures. In a sense, "disintegration" is the opposite process to the construction of a product measure.

Contenu 1 Motivation 2 Énoncé du théorème 3 Applications 3.1 Product spaces 3.2 Vector calculus 3.3 Conditional distributions 4 Voir également 5 References Motivation Consider the unit square in the Euclidean plane R2, S = [0, 1] × [0, 1]. Consider the probability measure μ defined on S by the restriction of two-dimensional Lebesgue measure λ2 to S. C'est-à-dire, the probability of an event E ⊆ S is simply the area of E. We assume E is a measurable subset of S.

Consider a one-dimensional subset of S such as the line segment Lx = {X} × [0, 1]. Lx has μ-measure zero; every subset of Lx is a μ-null set; since the Lebesgue measure space is a complete measure space, {displaystyle Esubseteq L_{X}implies mu (E)=0.} While true, this is somewhat unsatisfying. It would be nice to say that μ "restricted to" Lx is the one-dimensional Lebesgue measure λ1, rather than the zero measure. The probability of a "two-dimensional" event E could then be obtained as an integral of the one-dimensional probabilities of the vertical "slices" E ∩ Lx: more formally, if μx denotes one-dimensional Lebesgue measure on Lx, alors {style d'affichage lui (E)=int _{[0,1]}dans _{X}(Ecap L_{X}),mathrm {ré} X} pour toute "nice" E ⊆ S. The disintegration theorem makes this argument rigorous in the context of measures on metric spaces.

Énoncé du théorème (Hereafter, P(X) will denote the collection of Borel probability measures on a topological space (X, J).) The assumptions of the theorem are as follows: Let Y and X be two Radon spaces (c'est à dire. a topological space such that every Borel probability measure on M is inner regular e.g. separable metric spaces on which every probability measure is a Radon measure). Let μ ∈ P(Oui). Let π : Y → X be a Borel-measurable function. Here one should think of π as a function to "disintegrate" Oui, in the sense of partitioning Y into {style d'affichage {pi ^{-1}(X) | xin X}} . Par exemple, for the motivating example above, on peut définir {style d'affichage pi ((un,b))=un,(un,b)dans [0,1]fois [0,1]} , which gives that {displaystyle pi ^{-1}(un)=atimes [0,1]} , a slice we want to capture. Laisser {style d'affichage non } ∈ P(X) be the pushforward measure ν = π∗(m) = μ ∘ π−1. This measure provides the distribution of x (which corresponds to the events {displaystyle pi ^{-1}(X)} ).

The conclusion of the theorem: There exists a {style d'affichage non } -almost everywhere uniquely determined family of probability measures {μx}x∈X ⊆ P(Oui), which provides a "disintegration" de {style d'affichage lui } dans {style d'affichage {dans _{X}}_{xin X}} , tel que: the function {displaystyle xmapsto mu _{X}} is Borel measurable, dans le sens où {displaystyle xmapsto mu _{X}(B)} is a Borel-measurable function for each Borel-measurable set B ⊆ Y; μx "lives on" the fiber π−1(X): pour {style d'affichage non } -almost all x ∈ X, {style d'affichage lui _{X}la gauche(Ysetminus pi ^{-1}(X)droit)=0,} and so μx(E) = μx(E ∩ π−1(X)); for every Borel-measurable function f : Y → [0, ∞], {style d'affichage entier _{Oui}F(y),mathrm {ré} dans (y)=int _{X}entier _{pi ^{-1}(X)}F(y),mathrm {ré} dans _{X}(y)mathrm {ré} nu (X).} En particulier, for any event E ⊆ Y, taking f to be the indicator function of E,[1] {style d'affichage lui (E)=int _{X}dans _{X}la gauche(Eright),mathrm {ré} nu (X).} Applications Product spaces This section needs additional citations for verification. Aidez-nous à améliorer cet article en ajoutant des citations à des sources fiables. Le matériel non sourcé peut être contesté et supprimé. (Peut 2022) (Découvrez comment et quand supprimer ce modèle de message) The original example was a special case of the problem of product spaces, to which the disintegration theorem applies.

When Y is written as a Cartesian product Y = X1 × X2 and πi : Y → Xi is the natural projection, then each fibre π1−1(x1) can be canonically identified with X2 and there exists a Borel family of probability measures {style d'affichage {dans _{X_{1}}}_{X_{1}in X_{1}}} in P(X2) (lequel est (p1)∗(m)-almost everywhere uniquely determined) tel que {displaystyle mu =int _{X_{1}}dans _{X_{1}},mu left(pi _{1}^{-1}(mathrm {ré} X_{1})droit)=int _{X_{1}}dans _{X_{1}},mathrm {ré} (pi _{1})_{*}(dans )(X_{1}),} which is in particular[clarification nécessaire] {style d'affichage entier _{X_{1}times X_{2}}F(X_{1},X_{2}),dans (mathrm {ré} X_{1},mathrm {ré} X_{2})=int _{X_{1}}la gauche(entier _{X_{2}}F(X_{1},X_{2})dans (mathrm {ré} X_{2}|X_{1})droit)mu left(pi _{1}^{-1}(mathrm {ré} X_{1})droit)} et {style d'affichage lui (Afois B)=int _{UN}mu left(B|X_{1}droit),mu left(pi _{1}^{-1}(mathrm {ré} X_{1})droit).} The relation to conditional expectation is given by the identities {nom de l'opérateur de style d'affichage {E} (F|pi _{1})(X_{1})=int _{X_{2}}F(X_{1},X_{2})dans (mathrm {ré} X_{2}|X_{1}),} {style d'affichage lui (Afois B|pi _{1})(X_{1})=1_{UN}(X_{1})cdot mu (B|X_{1}).} Vector calculus The disintegration theorem can also be seen as justifying the use of a "restricted" measure in vector calculus. Par exemple, in Stokes' theorem as applied to a vector field flowing through a compact surface Σ ⊂ R3, it is implicit that the "correct" measure on Σ is the disintegration of three-dimensional Lebesgue measure λ3 on Σ, and that the disintegration of this measure on ∂Σ is the same as the disintegration of λ3 on ∂Σ.[2] Conditional distributions The disintegration theorem can be applied to give a rigorous treatment of conditional probability distributions in statistics, while avoiding purely abstract formulations of conditional probability.[3] See also Ionescu-Tulcea theorem Joint probability distribution – Type of probability distribution Copula (statistiques) Conditional expectation – Expected value of a random variable given that certain conditions are known to occur Regular conditional probability References ^ Dellacherie, C; Meyer, P.-A. (1978). Probabilities and Potential. Études de mathématiques en Hollande du Nord. Amsterdam: Hollande du Nord. ISBN 0-7204-0701-X. ^ Ambrosio, L, Gigli, N. & Savaré, g. (2005). Gradient Flows in Metric Spaces and in the Space of Probability Measures. ETH Zürich, Birkhäuser Verlag, Bâle. ISBN 978-3-7643-2428-5. ^ Chang, J.T.; Pollard, ré. (1997). "Conditioning as disintegration" (PDF). Statistica Neerlandica. 51 (3): 287. CiteSeerX 10.1.1.55.7544. est ce que je:10.1111/1467-9574.00056. S2CID 16749932. hide vte Measure theory Basic concepts Absolute continuityLebesgue integrationLp spacesMeasureMeasure space Probability spaceMeasurable space/function Sets Almost everywhereBorel setCarathéodory's criterionConvergence in measure -systemEssential range infimum/supremumLocally measurableπ-systemσ-algebraNon-measurable set Vitali setNull setSupportTransverse measure Types of Measures BaireBanachBesovBorelComplexCompleteContent(Logarithmically) ConvexDiscreteFiniteInner(Quasi-) InvariantLocally finiteMaximisingMetric outerOuterPerfectPre-measure(Sub-) ProbabilityProjection-valuedRadonRandomRegular Borel regularInner regularOuter regularSaturatedSet functionσ-finites-finiteSignedSingularSpectralStrictly positiveTightVector Particular measures CountingDiracEulerGaussianHaarHarmonicHausdorffIntensityLebesgueLogarithmicProductPushforwardSpherical measureTangentTrivialYoung Main results Carathéodory's extension theoremConvergence theorems DominatedMonotoneVitaliDecomposition theorems HahnJordanEgorov'sFatou's lemmaFubini'sHölder's inequalityMinkowski inequalityRadon–NikodymRiesz–Markov–Kakutani representation theorem Other results Disintegration theoremLebesgue's density theoremLebesgue differentiation theoremSard's theorem Applications Probability theoryReal analysisSpectral theory Categories: Théorèmes en théorie de la mesureThéorèmes de probabilité