Coarse-graining support: Support.Book.Ch04.Theorems.Expectations #
Imported from the Apache-2.0 CoarseGraining development at commit
c7ddd76c08ade64fed1b8d2ca51be14dfee8deb4.
Expectations and finite response moments #
This file is the public Chapter 4 surface for taking expectations of the
scalar response observables already exposed in CoarseObservables.
The surface is intentionally small: it names the observable, its expectation, the centered observable used in the manuscript, and the finite descendant average expectation theorem. Integrability remains an explicit theorem hypothesis; there is no extra moment carrier or section-local wrapper track.
Finite-dimensional matrix quadratic forms commute with entrywise expectation under entrywise integrability.
Scalar response observable on a deterministic triadic cube.
Equations
Instances For
The scalar response observable is pointwise nonnegative.
Full coarse-block integrability makes the scalar response integrable.
The law carrier supplies the a.s. elliptic support and the deterministic
ResponseJ = Mu(-p,q) - p·q identity; the only remaining analytic input is
integrability of the finite-dimensional coarse block.
Expected scalar response expressed through the annealed coarse block
matrix, with the stochastic hypotheses packaged in RestrictionLawCarrier.
This is the note-facing Chapter 4 source identity: Ch5 should call this rather than passing deterministic coarse-data witnesses.
Centered scalar response observable on a deterministic triadic cube.
Equations
- HCPolySupport.Book.Ch04.restrictionCenteredResponseJObservableCubeSet Q p q p0 q0 a = HCPolySupport.Book.Ch04.restrictionResponseJObservableCubeSet Q p q a - 1 / 2 * HCPolySupport.vecDot p0 q0
Instances For
Annealed scalar response on a deterministic triadic cube.
Equations
Instances For
Expected scalar response expressed through the annealed coarse block matrix.
The only stochastic hypotheses are the deterministic coarse-data identity almost surely and integrability of the full coarse block. Scalarization is not used here; Ch5 gets its scalar formula by specializing this source identity.
Annealed finite descendant average of scalar responses.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Difference of two annealed scalar responses, the basic τ-type quantity.
Equations
- HCPolySupport.Book.Ch04.tauResponseJCubeSet P Qchild Qparent p q = HCPolySupport.Book.Ch04.expectedResponseJCubeSet P Qchild p q - HCPolySupport.Book.Ch04.expectedResponseJCubeSet P Qparent p q
Instances For
Finite descendant averages preserve integrability.
Finite descendant averages commute with expectation under childwise integrability.
Finite descendant averages preserve MemLp.
Finite descendant averages of response observables are integrable if the child responses are integrable.
Finite descendant averages of response observables are in L^r if the
child responses are in L^r.
Centering by a deterministic scalar preserves integrability.
Centering by a deterministic scalar preserves MemLp under a finite
measure.
The integral of the centered response is the annealed response minus the deterministic centering scalar.
Finite descendant response averages commute with expectation, assuming childwise integrability.
The named response observable is a.e.-measurable under a law carrier.
The named response observable is a.e.-strongly-measurable under a law carrier.
Centered response observables are a.e.-measurable under a law carrier.
Centered response observables are a.e.-strongly-measurable under a law carrier.
Finite descendant averages of the named response observables are a.e.-measurable under a law carrier.
Finite descendant averages of the named response observables are a.e.-strongly-measurable under a law carrier.