Are ARIMA neural network hybrids better than single models?

Hybrid methods comprising autoregressive integrated moving average (ARIMA) and neural network models are generally favored against single neural network and single ARIMA models in the literature. The benefits of such methods appear to be substantial especially when dealing with non-stationary series...

Full description

Saved in:
Bibliographic Details
Main Authors: Taskaya-Temizel, T., Ahmad, K.
Format: Conference Proceeding
Language:English
Subjects:
Benchmark testing
Chaos
Computer networks
Economic forecasting
Electronic mail
Feedforward neural networks
Merging
Neural networks
Piecewise linear techniques
Statistics
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 3197 vol. 5
container_issue
container_start_page 3192
container_title
container_volume 5
creator Taskaya-Temizel, T.
Ahmad, K.
description Hybrid methods comprising autoregressive integrated moving average (ARIMA) and neural network models are generally favored against single neural network and single ARIMA models in the literature. The benefits of such methods appear to be substantial especially when dealing with non-stationary series: nonstationary linear component can be modeled using ARIMA and nonlinear component using neural networks. Our studies suggest that the use of a nonlinear component may degenerate the performance of such hybrids and that a simpler hybrid comprising linear AR model with a TDNN outperforms the more complex hybrid in tests on benchmark economic and financial time series.
doi_str_mv 10.1109/IJCNN.2005.1556438
format conference_proceeding
fullrecord <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_1556438</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>1556438</ieee_id><sourcerecordid>1556438</sourcerecordid><originalsourceid>FETCH-LOGICAL-i175t-cfa50578b7917584b42234ca6d92e35c5cab27aa33e29c4561beef59f23a41d63</originalsourceid><addsrcrecordid>eNo1j11LwzAUhoMf4Db9A3qTP9B6kpM0DV5IKVMrc4Lo9UjaU1ftOkkqsn_vwHn18PLACw9jlwJSIcBeV4_lcplKAJ0KrTOF-RGbSJGJRCkwx2wKJge0oHJ58i_Q4hmbxvgBINFanLCbIhAvXqqngg_0HVy_x_izDZ98vfOhayL3NI4U-Lh2A4_d8N4T32wb6uPtOTttXR_p4sAZe7ubv5YPyeL5viqLRdIJo8ekbp0GbXJv7H7nyispUdUua6wk1LWunZfGOUSStlY6E56o1baV6JRoMpyxq7_fjohWX6HbuLBbHaLxFz__SGM</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Are ARIMA neural network hybrids better than single models?</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Taskaya-Temizel, T. ; Ahmad, K.</creator><creatorcontrib>Taskaya-Temizel, T. ; Ahmad, K.</creatorcontrib><description>Hybrid methods comprising autoregressive integrated moving average (ARIMA) and neural network models are generally favored against single neural network and single ARIMA models in the literature. The benefits of such methods appear to be substantial especially when dealing with non-stationary series: nonstationary linear component can be modeled using ARIMA and nonlinear component using neural networks. Our studies suggest that the use of a nonlinear component may degenerate the performance of such hybrids and that a simpler hybrid comprising linear AR model with a TDNN outperforms the more complex hybrid in tests on benchmark economic and financial time series.</description><identifier>ISSN: 2161-4393</identifier><identifier>ISBN: 0780390482</identifier><identifier>ISBN: 9780780390485</identifier><identifier>EISSN: 2161-4407</identifier><identifier>DOI: 10.1109/IJCNN.2005.1556438</identifier><language>eng</language><publisher>IEEE</publisher><subject>Benchmark testing ; Chaos ; Computer networks ; Economic forecasting ; Electronic mail ; Feedforward neural networks ; Merging ; Neural networks ; Piecewise linear techniques ; Statistics</subject><ispartof>Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005, 2005, Vol.5, p.3192-3197 vol. 5</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/1556438$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,777,781,786,787,2052,4036,4037,27906,54536,54901,54913</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/1556438$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Taskaya-Temizel, T.</creatorcontrib><creatorcontrib>Ahmad, K.</creatorcontrib><title>Are ARIMA neural network hybrids better than single models?</title><title>Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005</title><addtitle>IJCNN</addtitle><description>Hybrid methods comprising autoregressive integrated moving average (ARIMA) and neural network models are generally favored against single neural network and single ARIMA models in the literature. The benefits of such methods appear to be substantial especially when dealing with non-stationary series: nonstationary linear component can be modeled using ARIMA and nonlinear component using neural networks. Our studies suggest that the use of a nonlinear component may degenerate the performance of such hybrids and that a simpler hybrid comprising linear AR model with a TDNN outperforms the more complex hybrid in tests on benchmark economic and financial time series.</description><subject>Benchmark testing</subject><subject>Chaos</subject><subject>Computer networks</subject><subject>Economic forecasting</subject><subject>Electronic mail</subject><subject>Feedforward neural networks</subject><subject>Merging</subject><subject>Neural networks</subject><subject>Piecewise linear techniques</subject><subject>Statistics</subject><issn>2161-4393</issn><issn>2161-4407</issn><isbn>0780390482</isbn><isbn>9780780390485</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2005</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNo1j11LwzAUhoMf4Db9A3qTP9B6kpM0DV5IKVMrc4Lo9UjaU1ftOkkqsn_vwHn18PLACw9jlwJSIcBeV4_lcplKAJ0KrTOF-RGbSJGJRCkwx2wKJge0oHJ58i_Q4hmbxvgBINFanLCbIhAvXqqngg_0HVy_x_izDZ98vfOhayL3NI4U-Lh2A4_d8N4T32wb6uPtOTttXR_p4sAZe7ubv5YPyeL5viqLRdIJo8ekbp0GbXJv7H7nyispUdUua6wk1LWunZfGOUSStlY6E56o1baV6JRoMpyxq7_fjohWX6HbuLBbHaLxFz__SGM</recordid><startdate>2005</startdate><enddate>2005</enddate><creator>Taskaya-Temizel, T.</creator><creator>Ahmad, K.</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>2005</creationdate><title>Are ARIMA neural network hybrids better than single models?</title><author>Taskaya-Temizel, T. ; Ahmad, K.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i175t-cfa50578b7917584b42234ca6d92e35c5cab27aa33e29c4561beef59f23a41d63</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2005</creationdate><topic>Benchmark testing</topic><topic>Chaos</topic><topic>Computer networks</topic><topic>Economic forecasting</topic><topic>Electronic mail</topic><topic>Feedforward neural networks</topic><topic>Merging</topic><topic>Neural networks</topic><topic>Piecewise linear techniques</topic><topic>Statistics</topic><toplevel>online_resources</toplevel><creatorcontrib>Taskaya-Temizel, T.</creatorcontrib><creatorcontrib>Ahmad, K.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Xplore (Online service)</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Taskaya-Temizel, T.</au><au>Ahmad, K.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Are ARIMA neural network hybrids better than single models?</atitle><btitle>Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005</btitle><stitle>IJCNN</stitle><date>2005</date><risdate>2005</risdate><volume>5</volume><spage>3192</spage><epage>3197 vol. 5</epage><pages>3192-3197 vol. 5</pages><issn>2161-4393</issn><eissn>2161-4407</eissn><isbn>0780390482</isbn><isbn>9780780390485</isbn><abstract>Hybrid methods comprising autoregressive integrated moving average (ARIMA) and neural network models are generally favored against single neural network and single ARIMA models in the literature. The benefits of such methods appear to be substantial especially when dealing with non-stationary series: nonstationary linear component can be modeled using ARIMA and nonlinear component using neural networks. Our studies suggest that the use of a nonlinear component may degenerate the performance of such hybrids and that a simpler hybrid comprising linear AR model with a TDNN outperforms the more complex hybrid in tests on benchmark economic and financial time series.</abstract><pub>IEEE</pub><doi>10.1109/IJCNN.2005.1556438</doi></addata></record>
fulltext fulltext_linktorsrc
identifier ISSN: 2161-4393
ispartof Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005, 2005, Vol.5, p.3192-3197 vol. 5
issn 2161-4393
2161-4407
language eng
recordid cdi_ieee_primary_1556438
source IEEE Electronic Library (IEL) Conference Proceedings
subjects Benchmark testing
Chaos
Computer networks
Economic forecasting
Electronic mail
Feedforward neural networks
Merging
Neural networks
Piecewise linear techniques
Statistics
title Are ARIMA neural network hybrids better than single models?
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-19T20%3A30%3A45IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Are%20ARIMA%20neural%20network%20hybrids%20better%20than%20single%20models?&rft.btitle=Proceedings.%202005%20IEEE%20International%20Joint%20Conference%20on%20Neural%20Networks,%202005&rft.au=Taskaya-Temizel,%20T.&rft.date=2005&rft.volume=5&rft.spage=3192&rft.epage=3197%20vol.%205&rft.pages=3192-3197%20vol.%205&rft.issn=2161-4393&rft.eissn=2161-4407&rft.isbn=0780390482&rft.isbn_list=9780780390485&rft_id=info:doi/10.1109/IJCNN.2005.1556438&rft_dat=%3Cieee_6IE%3E1556438%3C/ieee_6IE%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i175t-cfa50578b7917584b42234ca6d92e35c5cab27aa33e29c4561beef59f23a41d63%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=1556438&rfr_iscdi=true