Undecimated wavelet based autoregressive model for anchovy catches forecasting

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The aim of this paper is to find a model to forecast 1-month ahead monthly anchovy catches using un-decimated multi-scale stationary wavelet transform (USWT) combined with linear autoregressive (AR) method. The original monthly anchovy catches are decomposed into various sub-series employing USWT and then appropriate sub-series are used as inputs to the multi-scale autoregressive (MAR) model. The MAR's parameters are estimated using the regularized least squares (RLS) method. RLS based forecasting performance was evaluated using determination coefficient and shown that a 99% of the explained variance was captured with a reduced parsimony and high accuracy.

Original languageEnglish
Title of host publicationMICAI 2008
Subtitle of host publicationAdvances in Artificial Intelligence - 7th Mexican International Conference on Artificial Intelligence, Proceedings
Pages325-332
Number of pages8
DOIs
StatePublished - 5 Dec 2008
Event7th Mexican International Conference on Artificial Intelligence, MICAI 2008 - Atizapan de Zaragoza, Mexico
Duration: 27 Oct 200831 Oct 2008

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5317 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference7th Mexican International Conference on Artificial Intelligence, MICAI 2008
CountryMexico
CityAtizapan de Zaragoza
Period27/10/0831/10/08

Keywords

  • Autoregressive
  • Forecasting
  • Stationary wavelet transform

Fingerprint Dive into the research topics of 'Undecimated wavelet based autoregressive model for anchovy catches forecasting'. Together they form a unique fingerprint.

Cite this