Evolving stacked time series predictors with multiple window scales and sampling gaps

Zheng Rong Yang, Weiping Lu, Robert G. Harrison

Research output: Contribution to journalArticle

Abstract

We apply evolutionary programming to search for the optimal combination of stacked time series predictors with multiple window scales and sampling gaps. In this approach, the evolutionary process is ensured to proceed smoothly towards the optimal solution by using a control strategy based on the similarity level between the genotypes from two successive generations. Our experiments on both sunspots and S&P500 price index predictions demonstrate that this method significantly improves the prediction accuracy compared with the constrained least squared regression.

Original languageEnglish
Pages (from-to)203-211
Number of pages9
JournalNeural Processing Letters
Volume13
Issue number3
DOIs
Publication statusPublished - Jun 2001

Fingerprint

time series
sampling
prediction
sunspot
genotype
experiment
price
index
method

Keywords

  • Evolutionary programming
  • Forecasting
  • Neural networks
  • Stacking
  • Time series

Cite this

Yang, Zheng Rong ; Lu, Weiping ; Harrison, Robert G. / Evolving stacked time series predictors with multiple window scales and sampling gaps. In: Neural Processing Letters. 2001 ; Vol. 13, No. 3. pp. 203-211.
@article{f002d69d2edc403cb147a5218a75bf67,
title = "Evolving stacked time series predictors with multiple window scales and sampling gaps",
abstract = "We apply evolutionary programming to search for the optimal combination of stacked time series predictors with multiple window scales and sampling gaps. In this approach, the evolutionary process is ensured to proceed smoothly towards the optimal solution by using a control strategy based on the similarity level between the genotypes from two successive generations. Our experiments on both sunspots and S&P500 price index predictions demonstrate that this method significantly improves the prediction accuracy compared with the constrained least squared regression.",
keywords = "Evolutionary programming, Forecasting, Neural networks, Stacking, Time series",
author = "Yang, {Zheng Rong} and Weiping Lu and Harrison, {Robert G.}",
year = "2001",
month = "6",
doi = "10.1023/A:1011392725041",
language = "English",
volume = "13",
pages = "203--211",
journal = "Neural Processing Letters",
issn = "1370-4621",
publisher = "Springer",
number = "3",

}

Evolving stacked time series predictors with multiple window scales and sampling gaps. / Yang, Zheng Rong; Lu, Weiping; Harrison, Robert G.

In: Neural Processing Letters, Vol. 13, No. 3, 06.2001, p. 203-211.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Evolving stacked time series predictors with multiple window scales and sampling gaps

AU - Yang, Zheng Rong

AU - Lu, Weiping

AU - Harrison, Robert G.

PY - 2001/6

Y1 - 2001/6

N2 - We apply evolutionary programming to search for the optimal combination of stacked time series predictors with multiple window scales and sampling gaps. In this approach, the evolutionary process is ensured to proceed smoothly towards the optimal solution by using a control strategy based on the similarity level between the genotypes from two successive generations. Our experiments on both sunspots and S&P500 price index predictions demonstrate that this method significantly improves the prediction accuracy compared with the constrained least squared regression.

AB - We apply evolutionary programming to search for the optimal combination of stacked time series predictors with multiple window scales and sampling gaps. In this approach, the evolutionary process is ensured to proceed smoothly towards the optimal solution by using a control strategy based on the similarity level between the genotypes from two successive generations. Our experiments on both sunspots and S&P500 price index predictions demonstrate that this method significantly improves the prediction accuracy compared with the constrained least squared regression.

KW - Evolutionary programming

KW - Forecasting

KW - Neural networks

KW - Stacking

KW - Time series

UR - http://www.scopus.com/inward/record.url?scp=0035362641&partnerID=8YFLogxK

U2 - 10.1023/A:1011392725041

DO - 10.1023/A:1011392725041

M3 - Article

VL - 13

SP - 203

EP - 211

JO - Neural Processing Letters

JF - Neural Processing Letters

SN - 1370-4621

IS - 3

ER -