TheoremBase

Properties of the Total Variation: a Finite Measure, the Jordan Decomposition, Linear Combinations, the Total Variation Norm, and Integrability

The total variation of a finite signed measure is a finite measure equal to the sum of the Jordan parts and dominating the measure; it is subadditive and homogeneous under setwise linear combinations, so the total variation norm makes the finite signed measures a real normed space; nonnegative combinations of finite measures are finite measures with integrals combining accordingly; and a measurable function is integrable for the total variation exactly when it is integrable for both Jordan parts.

Statement

In the setting of Radon Measures on Locally Compact Metric Spaces: Standing Notation, let (X,F)(X,\mathcal{F}) be a measurable space, let ν\nu and ν′\nu' be finite signed measures on (X,F)(X,\mathcal{F}), and let c∈Rc\in\mathbb{R}. Let ∣ν∣|\nu| be the total variation and ∥ν∥\|\nu\| the total variation norm of ν\nu, and let ν+\nu^{+} and ν−\nu^{-} be its Jordan decomposition, formed with a Hahn set for ν\nu.

A finite measure. ∣ν∣|\nu| is a finite measure on (X,F)(X,\mathcal{F}).

Jordan decomposition. For every Hahn set PP for ν\nu and every A∈FA\in\mathcal{F},

∣ν∣(A)=ν+(A)+ν−(A)=ν(A∩P)−ν(A∖P),|\nu|(A)=\nu^{+}(A)+\nu^{-}(A)=\nu(A\cap P)-\nu(A\setminus P),

and ν+(A)≤∣ν∣(A)\nu^{+}(A)\le|\nu|(A) and ν−(A)≤∣ν∣(A)\nu^{-}(A)\le|\nu|(A).

Bound. For every A∈FA\in\mathcal{F},

∣ν(A)∣≤∣ν∣(A)≤∥ν∥.|\nu(A)|\le|\nu|(A)\le\|\nu\|.

Nonnegative signed measures. Suppose that ν(A)≥0\nu(A)\ge0 for every A∈FA\in\mathcal{F}. Then ν\nu is a finite measure on (X,F)(X,\mathcal{F}), XX is a Hahn set for ν\nu, and ∣ν∣(A)=ν+(A)=ν(A)|\nu|(A)=\nu^{+}(A)=\nu(A) and ν−(A)=0\nu^{-}(A)=0 for every A∈FA\in\mathcal{F}. Moreover every f∈L1(ν)f\in\mathcal{L}^{1}(\nu) lies in L1(ν+)\mathcal{L}^{1}(\nu^{+}) and in L1(ν−)\mathcal{L}^{1}(\nu^{-}), and

∫f dν+−∫f dν−=∫f dν.\int f\,d\nu^{+}-\int f\,d\nu^{-}=\int f\,d\nu .

Linear combinations. ν+ν′\nu+\nu' and cνc\nu are finite signed measures on (X,F)(X,\mathcal{F}).

Total variation of linear combinations. For every A∈FA\in\mathcal{F},

∣ν+ν′∣(A)≤∣ν∣(A)+∣ν′∣(A),∣cν∣(A)=∣c∣ ∣ν∣(A).|\nu+\nu'|(A)\le|\nu|(A)+|\nu'|(A),\qquad|c\nu|(A)=|c|\,|\nu|(A).

The total variation norm. M(X,F)M(X,\mathcal{F}) is a real vector space under the setwise operations, with zero vector the zero function on F\mathcal{F}, and ν↦∥ν∥\nu\mapsto\|\nu\| is a norm on it, so that (M(X,F),∥⋅∥)(M(X,\mathcal{F}),\|\cdot\|) is a real normed space. In particular ∥ν∥=0\|\nu\|=0 if and only if ν(A)=0\nu(A)=0 for every A∈FA\in\mathcal{F}.

Integrability. Let f:X→Rf:X\to\mathbb{R} be measurable. Then, for the integrals of ∣f∣∈M+|f|\in\mathcal{M}^{+},

∫∣f∣ d∣ν∣=∫∣f∣ dν++∫∣f∣ dν−,\int|f|\,d|\nu|=\int|f|\,d\nu^{+}+\int|f|\,d\nu^{-},

and f∈L1(∣ν∣)f\in\mathcal{L}^{1}(|\nu|) if and only if f∈L1(ν+)f\in\mathcal{L}^{1}(\nu^{+}) and f∈L1(ν−)f\in\mathcal{L}^{1}(\nu^{-}).

Proofs

Log in to submit a proof.

Loading...

Citations

Loading…

Dependencies

Loading…

Related

0 relations

Curated associations between results. These are editable and subjective — they do not replace the dependency graph, which is derived from the references in the text.

No relations recorded yet.

Comments

Log in to comment.

Loading…