Approximation of L₂-processes by Gaussian processes
Let \(T\) be an ergodic transformation of a nonatomic probability space, \(f\) an \(L^2\)-function and \(K\geq 1\) an integer. It is shown that there is another \(L^2\)-function \(g\), such that the joint distribution of \(T^ig\) \((1\leq i\leq K)\) is nearly normal, and such that the corresponding inner products \((T^i f, T^jf)\) and \((T^i g,T^jg)\) are nearly the same for \(1\leq i\), \(j\leq K\). This result is used to give a simple transparent proof of an important special case of a theorem of \textit{M. A. Akcoglu}, \textit{D. M. Ha} and \textit{R. L. Jones} [Can. J. Math. 49, No. 1, 3-23 (1997; Zbl 0870.28007)], refining the celebrated Bourgain entropy theorem.
- An application of the Neyman-Pearson lemma to Gaussian processes
- scientific article; zbMATH DE number 1603939 (Why is no real title available?)
- Optimal norm approximation in ergodic theory
- scientific article; zbMATH DE number 5520755 (Why is no real title available?)
- Sweeping Out Properties of Operator Sequences
- Reconstruction of two approximation processes in order to reproduce e^ax and e^2ax, a>0
This page was built for publication: Approximation of \(L_2\)-processes by Gaussian processes
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5906918)
