Supremum of a process in terms of trees

DOI10.1007/978-3-540-36428-3_12MaRDI QIDQ4424447zbMATH OpenFDO

Authors Olivier Guédon, Artem Zvavitch

Publication date 2003

Published in Lecture Notes in Mathematics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/978-3-540-36428-3_12

zbMATH Keywords

Gaussian processesset-valued treessupremum

Mathematics Subject Classification ID

General theory of stochastic processes (60G07) Gaussian processes (60G15) Stochastic processes (60G99)

Review text

Let \((T, d)\) be a compact metric space. For any real-valued stochastic process \(X_t\), define \[ E \sup_{t\in T}\, X_t= \sup\Biggl\{ E\sup_{t\in T_f}\, X_t,\;T_f\text{ finite subset of }T\Biggr\}. \] If \(X_t\) satisfies some concentration inequality (in particular if \(X_t\) is Gaussian), the authors find an upper and a lower bound of \(E\sup_{t\in T}\,X_t\) in terms of set-valued trees in \((T, d)\). More precisely they introduce the notions of labelled-covering and packing trees in \((T, d)\), of a size for each of them, express the upper and lower bounds in terms of these sizes, and prove furthermore that these two quantities are equivalent. For the entire collection see [Zbl 1009.00019].

Recommendations

• On the supremum of certain families of stochastic processes
• The Supremum of Some Canonical Processes
• Upper and lower bounds for stochastic processes. Modern methods and classical problems
• Random Gaussian sums on trees
• Majorizing measures on metric spaces

Cited in

(3)

• Cover times, blanket times, and majorizing measures
• Majorizing measures for the optimizer
• Random martingales and localization of maximal inequalities

This page was built for publication: Supremum of a process in terms of trees

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4424447)