Universal linear least-squares prediction in the presence of noise

Georg C. Zeitler, Andrew Carl Singer

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Universal linear least squares prediction of real-valued bounded individual sequences in the presence of additive bounded noise is considered. It is shown that there is a sequential predictor observing noisy samples of the sequence to be predicted only, whose loss in terms of the noise-free sequence is asymptotically as small as that of the best batch predictor out of the class of all linear predictors with knowledge of the entire noisy sequence in advance.

Original languageEnglish (US)
Title of host publication2007 IEEE/SP 14th Workshop on Statistical Signal Processing, SSP 2007, Proceedings
Pages611-614
Number of pages4
DOIs
StatePublished - 2007
Event2007 IEEE/SP 14th WorkShoP on Statistical Signal Processing, SSP 2007 - Madison, WI, United States
Duration: Aug 26 2007Aug 29 2007

Other

Other2007 IEEE/SP 14th WorkShoP on Statistical Signal Processing, SSP 2007
CountryUnited States
CityMadison, WI
Period8/26/078/29/07

Fingerprint

Additive noise

Keywords

  • Least squares
  • Linear
  • Noise
  • Prediction

ASJC Scopus subject areas

  • Signal Processing

Cite this

Zeitler, G. C., & Singer, A. C. (2007). Universal linear least-squares prediction in the presence of noise. In 2007 IEEE/SP 14th Workshop on Statistical Signal Processing, SSP 2007, Proceedings (pp. 611-614). [4301331] https://doi.org/10.1109/SSP.2007.4301331

Universal linear least-squares prediction in the presence of noise. / Zeitler, Georg C.; Singer, Andrew Carl.

2007 IEEE/SP 14th Workshop on Statistical Signal Processing, SSP 2007, Proceedings. 2007. p. 611-614 4301331.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zeitler, GC & Singer, AC 2007, Universal linear least-squares prediction in the presence of noise. in 2007 IEEE/SP 14th Workshop on Statistical Signal Processing, SSP 2007, Proceedings., 4301331, pp. 611-614, 2007 IEEE/SP 14th WorkShoP on Statistical Signal Processing, SSP 2007, Madison, WI, United States, 8/26/07. https://doi.org/10.1109/SSP.2007.4301331
Zeitler GC, Singer AC. Universal linear least-squares prediction in the presence of noise. In 2007 IEEE/SP 14th Workshop on Statistical Signal Processing, SSP 2007, Proceedings. 2007. p. 611-614. 4301331 https://doi.org/10.1109/SSP.2007.4301331
Zeitler, Georg C. ; Singer, Andrew Carl. / Universal linear least-squares prediction in the presence of noise. 2007 IEEE/SP 14th Workshop on Statistical Signal Processing, SSP 2007, Proceedings. 2007. pp. 611-614
@inproceedings{4683e97fd326446c8d997b5f8f432b94,
title = "Universal linear least-squares prediction in the presence of noise",
abstract = "Universal linear least squares prediction of real-valued bounded individual sequences in the presence of additive bounded noise is considered. It is shown that there is a sequential predictor observing noisy samples of the sequence to be predicted only, whose loss in terms of the noise-free sequence is asymptotically as small as that of the best batch predictor out of the class of all linear predictors with knowledge of the entire noisy sequence in advance.",
keywords = "Least squares, Linear, Noise, Prediction",
author = "Zeitler, {Georg C.} and Singer, {Andrew Carl}",
year = "2007",
doi = "10.1109/SSP.2007.4301331",
language = "English (US)",
isbn = "142441198X",
pages = "611--614",
booktitle = "2007 IEEE/SP 14th Workshop on Statistical Signal Processing, SSP 2007, Proceedings",

}

TY - GEN

T1 - Universal linear least-squares prediction in the presence of noise

AU - Zeitler, Georg C.

AU - Singer, Andrew Carl

PY - 2007

Y1 - 2007

N2 - Universal linear least squares prediction of real-valued bounded individual sequences in the presence of additive bounded noise is considered. It is shown that there is a sequential predictor observing noisy samples of the sequence to be predicted only, whose loss in terms of the noise-free sequence is asymptotically as small as that of the best batch predictor out of the class of all linear predictors with knowledge of the entire noisy sequence in advance.

AB - Universal linear least squares prediction of real-valued bounded individual sequences in the presence of additive bounded noise is considered. It is shown that there is a sequential predictor observing noisy samples of the sequence to be predicted only, whose loss in terms of the noise-free sequence is asymptotically as small as that of the best batch predictor out of the class of all linear predictors with knowledge of the entire noisy sequence in advance.

KW - Least squares

KW - Linear

KW - Noise

KW - Prediction

UR - http://www.scopus.com/inward/record.url?scp=47849131916&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=47849131916&partnerID=8YFLogxK

U2 - 10.1109/SSP.2007.4301331

DO - 10.1109/SSP.2007.4301331

M3 - Conference contribution

AN - SCOPUS:47849131916

SN - 142441198X

SN - 9781424411986

SP - 611

EP - 614

BT - 2007 IEEE/SP 14th Workshop on Statistical Signal Processing, SSP 2007, Proceedings

ER -