On estimation of maximum values of nonparametric signals for random fields

MaRDI QIDQ2739829zbMATH OpenFDO

Authors A. A. Kolosov

Publication date 16 September 2001

Published in Prykladna Statystyka. Aktuarna ta Finansova Matematyka (Search for Journal in Brave)

zbMATH Keywords

minimax risksignal processing

Mathematics Subject Classification ID

Stochastic ordinary differential equations (aspects of stochastic analysis) (60H10) Nonparametric estimation (62G05) Asymptotic properties of nonparametric inference (62G20) Random fields; image analysis (62M40)

Review text

Let the random field \(X_\varepsilon(t)\) be described by the model NEWLINE\[NEWLINEdX_\varepsilon(t)=S(t)dt+\varepsilon dW(t),NEWLINE\]NEWLINE where \(W(t)\) is a Wiener field observed on \(t\in[0,1]^n\). The problem is to estimate \(\sup_{t\in[0,1]^n}S(t)\) when \(S\) satisfies some Hölder conditions. A lower bound for the minimax risk is obtained and an asymptotically efficient estimator (as \(\varepsilon\to 0\)) is constructed which is the sup of a kernel smoothed observed trajectory of \(X_\varepsilon\).

Recommendations

• scientific article; zbMATH DE number 482625
• The distribution of maxima of approximately Gaussian random fields
• Estimation of the maximum value of a signal in Gaussian white noise
• scientific article; zbMATH DE number 3923790
• Оценки распределения максимума случайного поля

Cited in

(1)

• scientific article; zbMATH DE number 482625 (Why is no real title available?)

This page was built for publication: On estimation of maximum values of nonparametric signals for random fields

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2739829)