Characterization of Integrable Functions
Theorem
Let $\struct {X, \Sigma, \mu}$ be a measure space.
Let $f: X \to \overline \R, f \in \MM_{\overline \R}$ be a $\Sigma$-measurable function.
Then the following are equivalent:
- $(1): \quad f \in \map {\LL_{\overline \R} } \mu$, that is, $f$ is $\mu$-integrable.
- $(2): \quad$ The positive and negative parts $f^+$ and $f^-$ are $\mu$-integrable.
- $(3): \quad$ The absolute value $\size f$ of $f$ is $\mu$-integrable.
- $(4): \quad$ There exists an $\mu$-integrable function $g: X \to \overline \R$ such that $\size f \le g$ pointwise.
![]() | This page has been identified as a candidate for refactoring of advanced complexity. In particular: Several results here, making specific referencing imprecise Until this has been finished, please leave {{Refactor}} in the code.
New contributors: Refactoring is a task which is expected to be undertaken by experienced editors only. Because of the underlying complexity of the work needed, it is recommended that you do not embark on a refactoring task until you have become familiar with the structural nature of pages of $\mathsf{Pr} \infty \mathsf{fWiki}$.To discuss this page in more detail, feel free to use the talk page. When this work has been completed, you may remove this instance of {{Refactor}} from the code. |
Proof
We prove the whole cycle of implications:
- $(1) \implies (2) \quad$ by definition of $(1)$
- $(2) \implies (3)\quad$ because $\size f = f^+ + f^-$ and Integral of Positive Measurable Function is Additive
- $(3) \implies (4)\quad$ because $g:= \size f$ exists
It remains to demonstrate $(4) \implies (1)$.
Let $f \in \MM_{\overline \R}$ and $g$ according to $(4)$.
Then:
- $f = f^+ - f^-$
where $f^+$ is the positive and $f^-$ is the negative part of $f$.
We have that $f^+$ and $f^-$ are positive and measurable.
Let $f^0$ stand for either $f^+$ or $f^-$.
We have that:
- $\size f = f^+ + f^-$
Therefore:
- $f^0 \le \size f \le g$
It is to be shown that the Integral of Positive Measurable Function of $f^0$ exists and is finite.
Let $\EE^+$ and $\map {I_\mu} h$ be defined as in Integral of Positive Measurable Function.
Then:
- $\forall h \in \EE^+$: $h \le f^0 \implies h \le g$
Hence:
- $\set {h: h \le f^0, h \in \EE^+} \subseteq \set {h: h \le g, h \in \EE^+}$
- $\ds \int f^0 \rd \mu := \sup \set {\map {I_\mu} h: h \le f^0, h \in \EE^+} \le \sup \set {\map {I_\mu} h: h \le g, h \in \EE^+} < \infty$
We have that the integrals for $f^+$ and $f^-$ both are finite.
Therefore $f$ is $\mu$-integrable according to definition.
$\blacksquare$
![]() | This article needs to be linked to other articles. In particular: The steps in the above need to be demonstrated and explained by linking to appropriate results on $\mathsf{Pr} \infty \mathsf{fWiki}$ As it stands, none of the above can be followed without knowing all about this subject. You can help $\mathsf{Pr} \infty \mathsf{fWiki}$ by adding these links. To discuss this page in more detail, feel free to use the talk page. When this work has been completed, you may remove this instance of {{MissingLinks}} from the code. |
.
Sources
- 2005: René L. Schilling: Measures, Integrals and Martingales ... (previous) ... (next): $10.3$