An ergodic theorem without invariant measure

Invariant Measure, the Recurrence Theorem, and the Ergodic Theorem

converge a.e. to a finite limit, where fs is the characteristic function of the set E. G. D. Birkhoff's Ergodic Theorem asserts this conclusion if T is measure-preserving, in the sense that m(T~1E)=m(E) for each measurable set E. The same conclusion can be asserted under somewhat more general circumstances. We shall say that T admits a finite, equivalent, invariant measure if there is a finite ...

Ergodic Theory of Markov Chains Admitting an Infinite Invariant Measure.

An Invariant Subspace Theorem

for every rational function ƒ with poles off K. In this note it is shown that any operator for which the spectrum is a spectral set has a nontrivial invariant subspace. In [6] von Neumann introduced the notion of spectral set and showed that if T has II T\\ = 1 then the closed unit disc, D"~, is a spectral set for T. For this reason any operator whose spectrum is a spectral set is called a von ...

Kingman's Subadditive Ergodic Theorem Kingman's Subadditive Ergodic Theorem

A simple proof of Kingman’s subadditive ergodic theorem is developed from a point of view which is conceptually algorithmic and which does not rely on either a maximal inequality or a combinatorial Riesz lemma.

An Ergodic Theorem for Stochastic Programming Problems?

To justify the use of sampling to solve stochastic programming problems one usually relies on a law of large numbers for random lsc (lower semicontinuous) functions when the samples come from independent, identical experiments. If the samples come from a stationary process, one can appeal to the ergodic theorem proved here. The proof relies on thèscalarization' of random lsc functions.