Lebesgue's decomposition theorem

From Wikipedia, the free encyclopedia
Jump to: navigation, search

In mathematics, more precisely in measure theory, Lebesgue's decomposition theorem[1][2][3] states that for every two σ-finite signed measures \mu and \nu on a measurable space (\Omega,\Sigma), there exist two σ-finite signed measures \nu_0 and \nu_1 such that:

These two measures are uniquely determined by \mu and \nu.

Refinement[edit]

Lebesgue's decomposition theorem can be refined in a number of ways.

First, the decomposition of the singular part of a regular Borel measure on the real line can be refined:[4]

\, \nu = \nu_{\mathrm{cont}} + \nu_{\mathrm{sing}} + \nu_{\mathrm{pp}}

where

  • νcont is the absolutely continuous part
  • νsing is the singular continuous part
  • νpp is the pure point part (a discrete measure).

Second, absolutely continuous measures are classified by the Radon–Nikodym theorem, and discrete measures are easily understood. Hence (singular continuous measures aside), Lebesgue decomposition gives a very explicit description of measures. The Cantor measure (the probability measure on the real line whose cumulative distribution function is the Cantor function) is an example of a singular continuous measure.

Related concepts[edit]

Lévy–Itō decomposition[edit]

The analogous decomposition for a stochastic processes is the Lévy–Itō decomposition: given a Lévy process X, it can be decomposed as a sum of three independent Lévy processes X=X^{(1)}+X^{(2)}+X^{(3)} where:

See also[edit]

Citations[edit]

  1. ^ (Halmos 1974, Section 32, Theorem C)
  2. ^ (Hewitt & Stromberg 1965, Chapter V, § 19, (19.42) Lebesgue Decomposition Theorem)
  3. ^ (Rudin 1974, Section 6.9, The Theorem of Lebesgue-Radon-Nikodym)
  4. ^ (Hewitt & Stromberg 1965, Chapter V, § 19, (19.61) Theorem)

References[edit]

This article incorporates material from Lebesgue decomposition theorem on PlanetMath, which is licensed under the Creative Commons Attribution/Share-Alike License.