On universal linear prediction of Gaussian data

Suleyman S. Kozat; Andrew C. Singer

doi:10.1109/ICASSP.2000.861846

On universal linear prediction of Gaussian data

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper, we derive some of the stochastic properties of a universal linear predictor, through analyses similar to those generally made in the adaptive signal processing literature. A. C. Singer et al. (see IEEE Trans. Signal Proc., vol.47, no.10, p.2685-2700, Oct. 1999) introduced a predictor whose sequentially accumulated mean squared error for any bounded individual sequence was shown to be as small as that for any linear predictor of order less than some maximum order m. For stationary Gaussian time series, we generalize these results, and remove the boundedness restriction. In this paper we show that the learning curve of this universal linear predictor is dominated by the learning curve of the best order predictor used in the algorithm.

Original language	English (US)
Title of host publication	Signal Processing Theory and Methods I
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	13-16
Number of pages	4
ISBN (Electronic)	0780362934
DOIs	https://doi.org/10.1109/ICASSP.2000.861846
State	Published - Jan 1 2000
Event	25th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2000 - Istanbul, Turkey Duration: Jun 5 2000 → Jun 9 2000

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	1
ISSN (Print)	1520-6149

Other

Other	25th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2000
Country	Turkey
City	Istanbul
Period	6/5/00 → 6/9/00

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP.2000.861846

Fingerprint Dive into the research topics of 'On universal linear prediction of Gaussian data'. Together they form a unique fingerprint.

Cite this

On universal linear prediction of Gaussian data. / Kozat, Suleyman S.; Singer, Andrew C.

Signal Processing Theory and Methods I. Institute of Electrical and Electronics Engineers Inc., 2000. p. 13-16 861846 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 1).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Kozat, SS & Singer, AC 2000, On universal linear prediction of Gaussian data. in Signal Processing Theory and Methods I., 861846, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 1, Institute of Electrical and Electronics Engineers Inc., pp. 13-16, 25th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2000, Istanbul, Turkey, 6/5/00. https://doi.org/10.1109/ICASSP.2000.861846

@inproceedings{085b45cd5fa546d6abe2968cf4bc5fd8,

title = "On universal linear prediction of Gaussian data",

abstract = "In this paper, we derive some of the stochastic properties of a universal linear predictor, through analyses similar to those generally made in the adaptive signal processing literature. A. C. Singer et al. (see IEEE Trans. Signal Proc., vol.47, no.10, p.2685-2700, Oct. 1999) introduced a predictor whose sequentially accumulated mean squared error for any bounded individual sequence was shown to be as small as that for any linear predictor of order less than some maximum order m. For stationary Gaussian time series, we generalize these results, and remove the boundedness restriction. In this paper we show that the learning curve of this universal linear predictor is dominated by the learning curve of the best order predictor used in the algorithm.",

author = "Kozat, {Suleyman S.} and Singer, {Andrew C.}",

year = "2000",

month = jan,

day = "1",

doi = "10.1109/ICASSP.2000.861846",

language = "English (US)",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "13--16",

booktitle = "Signal Processing Theory and Methods I",

address = "United States",

note = "25th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2000 ; Conference date: 05-06-2000 Through 09-06-2000",

}

TY - GEN

T1 - On universal linear prediction of Gaussian data

AU - Kozat, Suleyman S.

AU - Singer, Andrew C.

PY - 2000/1/1

Y1 - 2000/1/1

N2 - In this paper, we derive some of the stochastic properties of a universal linear predictor, through analyses similar to those generally made in the adaptive signal processing literature. A. C. Singer et al. (see IEEE Trans. Signal Proc., vol.47, no.10, p.2685-2700, Oct. 1999) introduced a predictor whose sequentially accumulated mean squared error for any bounded individual sequence was shown to be as small as that for any linear predictor of order less than some maximum order m. For stationary Gaussian time series, we generalize these results, and remove the boundedness restriction. In this paper we show that the learning curve of this universal linear predictor is dominated by the learning curve of the best order predictor used in the algorithm.

AB - In this paper, we derive some of the stochastic properties of a universal linear predictor, through analyses similar to those generally made in the adaptive signal processing literature. A. C. Singer et al. (see IEEE Trans. Signal Proc., vol.47, no.10, p.2685-2700, Oct. 1999) introduced a predictor whose sequentially accumulated mean squared error for any bounded individual sequence was shown to be as small as that for any linear predictor of order less than some maximum order m. For stationary Gaussian time series, we generalize these results, and remove the boundedness restriction. In this paper we show that the learning curve of this universal linear predictor is dominated by the learning curve of the best order predictor used in the algorithm.

UR - http://www.scopus.com/inward/record.url?scp=0033708397&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033708397&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2000.861846

DO - 10.1109/ICASSP.2000.861846

M3 - Conference contribution

AN - SCOPUS:0033708397

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 13

EP - 16

BT - Signal Processing Theory and Methods I

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 25th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2000

Y2 - 5 June 2000 through 9 June 2000

ER -