On-line tracking of a smooth regression function

Lev Goldentayer; Robert Liptser

doi:10.1007/s11203-004-8121-3

On-line tracking of a smooth regression function

Lev Goldentayer, Robert Liptser^*

^*Corresponding author for this work

Engineering General

Research output: Contribution to journal › Article › peer-review

Abstract

We construct an on-line estimator with equidistant design for tracking a smooth function from Stone-Ibragimov-Khasminskii's class. This estimator has the optimal convergence rate of risk to zero in sample size. The procedure for setting coefficients of the estimator is controlled by a single parameter and has a simple numerical solution. The off-line version of this estimator allows to eliminate a boundary layer. Simulation results are given.

Original language	English
Pages (from-to)	17-30
Number of pages	14
Journal	Statistical Inference for Stochastic Processes
Volume	9
Issue number	1
DOIs	https://doi.org/10.1007/s11203-004-8121-3
State	Published - May 2006

Keywords

Equidistant design
On-line tracking estimator

Access to Document

10.1007/s11203-004-8121-3

Cite this

@article{7c376e4e1cfc49079e0a37e9a689ddc0,

title = "On-line tracking of a smooth regression function",

abstract = "We construct an on-line estimator with equidistant design for tracking a smooth function from Stone-Ibragimov-Khasminskii's class. This estimator has the optimal convergence rate of risk to zero in sample size. The procedure for setting coefficients of the estimator is controlled by a single parameter and has a simple numerical solution. The off-line version of this estimator allows to eliminate a boundary layer. Simulation results are given.",

keywords = "Equidistant design, On-line tracking estimator",

author = "Lev Goldentayer and Robert Liptser",

note = "Funding Information: *This work is partially supported by a fellowship from the Yitzhak and Chaya Weinstein Research Institute for Signal Processing at Tel Aviv University. †Author for correspondence: Tel.: 972-0(3)-640-7195; Fax: 972(0)3-640-7095; e-mail: liptser@ eng.tau.ac.il",

year = "2006",

month = may,

doi = "10.1007/s11203-004-8121-3",

language = "אנגלית",

volume = "9",

pages = "17--30",

journal = "Statistical Inference for Stochastic Processes",

issn = "1387-0874",

publisher = "Springer Netherlands",

number = "1",

}

TY - JOUR

T1 - On-line tracking of a smooth regression function

AU - Goldentayer, Lev

AU - Liptser, Robert

N1 - Funding Information: *This work is partially supported by a fellowship from the Yitzhak and Chaya Weinstein Research Institute for Signal Processing at Tel Aviv University. †Author for correspondence: Tel.: 972-0(3)-640-7195; Fax: 972(0)3-640-7095; e-mail: liptser@ eng.tau.ac.il

PY - 2006/5

Y1 - 2006/5

N2 - We construct an on-line estimator with equidistant design for tracking a smooth function from Stone-Ibragimov-Khasminskii's class. This estimator has the optimal convergence rate of risk to zero in sample size. The procedure for setting coefficients of the estimator is controlled by a single parameter and has a simple numerical solution. The off-line version of this estimator allows to eliminate a boundary layer. Simulation results are given.

AB - We construct an on-line estimator with equidistant design for tracking a smooth function from Stone-Ibragimov-Khasminskii's class. This estimator has the optimal convergence rate of risk to zero in sample size. The procedure for setting coefficients of the estimator is controlled by a single parameter and has a simple numerical solution. The off-line version of this estimator allows to eliminate a boundary layer. Simulation results are given.

KW - Equidistant design

KW - On-line tracking estimator

UR - http://www.scopus.com/inward/record.url?scp=33646529198&partnerID=8YFLogxK

U2 - 10.1007/s11203-004-8121-3

DO - 10.1007/s11203-004-8121-3

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

AN - SCOPUS:33646529198

SN - 1387-0874

VL - 9

SP - 17

EP - 30

JO - Statistical Inference for Stochastic Processes

JF - Statistical Inference for Stochastic Processes

IS - 1

ER -

On-line tracking of a smooth regression function

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this