Financial Forecasting in Consumer Cyclicals with Economic Indicators and Tokenization

Krawczyk, Kasper; Tam, Gary; Archambault, Daniel

doi:10.7763/ijcte.2025.v17.1380

Journal article 428 views 23 downloads

Financial Forecasting in Consumer Cyclicals with Economic Indicators and Tokenization

Kasper Krawczyk, Gary Tam

, Daniel Archambault

International Journal of Computer Theory and Engineering, Volume: 17, Issue: 4, Pages: 179 - 188

Swansea University Authors: Kasper Krawczyk, Gary Tam

PDF | Version of Record

© 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY).
Download (2.32MB)

Check full text

DOI (Published version): 10.7763/ijcte.2025.v17.1380

Abstract

Machine learning in financial Time Series Forecasting (TSF) has a clear application in investment, where predicting stock price movements can inform investment strategies. The Transformer model has emerged as a powerful tool for this purpose, yet significant research gaps remain. Existing studies of...

Full description

Published in:	International Journal of Computer Theory and Engineering
ISSN:	1793-8201 2972-4511
Published:	IACSIT Press 2025
Online Access:	Check full text
URI:	https://cronfa.swan.ac.uk/Record/cronfa69661

first_indexed	2025-06-09T16:27:03Z
last_indexed	2025-10-31T08:45:39Z
id	cronfa69661
recordtype	SURis
fullrecord	<?xml version="1.0"?><rfc1807><datestamp>2025-10-29T15:14:53.4309430</datestamp><bib-version>v2</bib-version><id>69661</id><entry>2025-06-09</entry><title>Financial Forecasting in Consumer Cyclicals with Economic Indicators and Tokenization</title><swanseaauthors><author><sid>28171929b8b38037cf658e47f91683f1</sid><firstname>Kasper</firstname><surname>Krawczyk</surname><name>Kasper Krawczyk</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>e75a68e11a20e5f1da94ee6e28ff5e76</sid><ORCID>0000-0001-7387-5180</ORCID><firstname>Gary</firstname><surname>Tam</surname><name>Gary Tam</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2025-06-09</date><abstract>Machine learning in financial Time Series Forecasting (TSF) has a clear application in investment, where predicting stock price movements can inform investment strategies. The Transformer model has emerged as a powerful tool for this purpose, yet significant research gaps remain. Existing studies often focus on a set of stocks without modeling stock behavior within specific sectors. Notably, there is a lack of research on the Consumer Cyclicals (CC) industry, which includes sectors such as automotive, housing, entertainment, and retail. These industries are highly sensitive to economic conditions, making them crucial for understanding broader economic impacts on stock behavior. Furthermore, we observe that many existing works neglect broader economic contexts, which is particularly important for CC analysis due to its sensitivity to economic trends. Additionally, previous studies on financial Transformer models typically use model tokens as feature vectors of multiple variates at a single timestep. This approach may not adequately capture the important relationships between dataset variates for long-term economic trends. To this end, we present the first study on financial TSF for the CC sector along with economic data. To support the research, we propose the first public benchmark dataset for the CC sector, consisting of traditional stock price time series, technical indicators, and temporal data, enriched with economic indicators. Next, we introduce an alternative tokenization approach to enhance the model’s ability to capture long-term trends by preserving information about nonlinear dependencies between dataset variates. We hypothesize that this approach helps capture long-term signals more effectively. Through a comprehensive data ablation study and benchmark testing, we demonstrate that incorporating economic indicators improves the accuracy of longer-term predictions for the CC sector, and the new tokenization method enhances the performance of Transformer models. The dataset and code are made publicly available at: https://github.com/KasperKrawczyk/econtrans_dataset</abstract><type>Journal Article</type><journal>International Journal of Computer Theory and Engineering</journal><volume>17</volume><journalNumber>4</journalNumber><paginationStart>179</paginationStart><paginationEnd>188</paginationEnd><publisher>IACSIT Press</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>1793-8201</issnPrint><issnElectronic>2972-4511</issnElectronic><keywords>time series, forecasting, transformer, finance, stocks</keywords><publishedDay>23</publishedDay><publishedMonth>10</publishedMonth><publishedYear>2025</publishedYear><publishedDate>2025-10-23</publishedDate><doi>10.7763/ijcte.2025.v17.1380</doi><url/><notes>The dataset and code are made publicly available at: https://github.com/KasperKrawczyk/econtrans_dataset</notes><college>COLLEGE NANME</college><CollegeCode>COLLEGE CODE</CollegeCode><institution>Swansea University</institution><apcterm>Not Required</apcterm><funders/><projectreference/><lastEdited>2025-10-29T15:14:53.4309430</lastEdited><Created>2025-06-09T17:24:25.5566752</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Kasper</firstname><surname>Krawczyk</surname><order>1</order></author><author><firstname>Gary</firstname><surname>Tam</surname><orcid>0000-0001-7387-5180</orcid><order>2</order></author><author><firstname>Daniel</firstname><surname>Archambault</surname><orcid>0000-0003-4978-8479</orcid><order>3</order></author></authors><documents><document><filename>69661__35494__8400cac840544923996d9ceaf7e7e5a7.pdf</filename><originalFilename>69661.VOR.pdf</originalFilename><uploaded>2025-10-29T15:08:01.3512284</uploaded><type>Output</type><contentLength>2427810</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>© 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY).</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling	2025-10-29T15:14:53.4309430 v2 69661 2025-06-09 Financial Forecasting in Consumer Cyclicals with Economic Indicators and Tokenization 28171929b8b38037cf658e47f91683f1 Kasper Krawczyk Kasper Krawczyk true false e75a68e11a20e5f1da94ee6e28ff5e76 0000-0001-7387-5180 Gary Tam Gary Tam true false 2025-06-09 Machine learning in financial Time Series Forecasting (TSF) has a clear application in investment, where predicting stock price movements can inform investment strategies. The Transformer model has emerged as a powerful tool for this purpose, yet significant research gaps remain. Existing studies often focus on a set of stocks without modeling stock behavior within specific sectors. Notably, there is a lack of research on the Consumer Cyclicals (CC) industry, which includes sectors such as automotive, housing, entertainment, and retail. These industries are highly sensitive to economic conditions, making them crucial for understanding broader economic impacts on stock behavior. Furthermore, we observe that many existing works neglect broader economic contexts, which is particularly important for CC analysis due to its sensitivity to economic trends. Additionally, previous studies on financial Transformer models typically use model tokens as feature vectors of multiple variates at a single timestep. This approach may not adequately capture the important relationships between dataset variates for long-term economic trends. To this end, we present the first study on financial TSF for the CC sector along with economic data. To support the research, we propose the first public benchmark dataset for the CC sector, consisting of traditional stock price time series, technical indicators, and temporal data, enriched with economic indicators. Next, we introduce an alternative tokenization approach to enhance the model’s ability to capture long-term trends by preserving information about nonlinear dependencies between dataset variates. We hypothesize that this approach helps capture long-term signals more effectively. Through a comprehensive data ablation study and benchmark testing, we demonstrate that incorporating economic indicators improves the accuracy of longer-term predictions for the CC sector, and the new tokenization method enhances the performance of Transformer models. The dataset and code are made publicly available at: https://github.com/KasperKrawczyk/econtrans_dataset Journal Article International Journal of Computer Theory and Engineering 17 4 179 188 IACSIT Press 1793-8201 2972-4511 time series, forecasting, transformer, finance, stocks 23 10 2025 2025-10-23 10.7763/ijcte.2025.v17.1380 The dataset and code are made publicly available at: https://github.com/KasperKrawczyk/econtrans_dataset COLLEGE NANME COLLEGE CODE Swansea University Not Required 2025-10-29T15:14:53.4309430 2025-06-09T17:24:25.5566752 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Kasper Krawczyk 1 Gary Tam 0000-0001-7387-5180 2 Daniel Archambault 0000-0003-4978-8479 3 69661__35494__8400cac840544923996d9ceaf7e7e5a7.pdf 69661.VOR.pdf 2025-10-29T15:08:01.3512284 Output 2427810 application/pdf Version of Record true © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY). true eng https://creativecommons.org/licenses/by/4.0/
title	Financial Forecasting in Consumer Cyclicals with Economic Indicators and Tokenization
spellingShingle	Financial Forecasting in Consumer Cyclicals with Economic Indicators and Tokenization Kasper Krawczyk Gary Tam
title_short	Financial Forecasting in Consumer Cyclicals with Economic Indicators and Tokenization
title_full	Financial Forecasting in Consumer Cyclicals with Economic Indicators and Tokenization
title_fullStr	Financial Forecasting in Consumer Cyclicals with Economic Indicators and Tokenization
title_full_unstemmed	Financial Forecasting in Consumer Cyclicals with Economic Indicators and Tokenization
title_sort	Financial Forecasting in Consumer Cyclicals with Economic Indicators and Tokenization
author_id_str_mv	28171929b8b38037cf658e47f91683f1 e75a68e11a20e5f1da94ee6e28ff5e76
author_id_fullname_str_mv	28171929b8b38037cf658e47f91683f1_*_Kasper Krawczyk e75a68e11a20e5f1da94ee6e28ff5e76_*_Gary Tam
author	Kasper Krawczyk Gary Tam
author2	Kasper Krawczyk Gary Tam Daniel Archambault
format	Journal article
container_title	International Journal of Computer Theory and Engineering
container_volume	17
container_issue	4
container_start_page	179
publishDate	2025
institution	Swansea University
issn	1793-8201 2972-4511
doi_str_mv	10.7763/ijcte.2025.v17.1380
publisher	IACSIT Press
college_str	Faculty of Science and Engineering
hierarchytype
hierarchy_top_id	facultyofscienceandengineering
hierarchy_top_title	Faculty of Science and Engineering
hierarchy_parent_id	facultyofscienceandengineering
hierarchy_parent_title	Faculty of Science and Engineering
department_str	School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str	1
active_str	0
description	Machine learning in financial Time Series Forecasting (TSF) has a clear application in investment, where predicting stock price movements can inform investment strategies. The Transformer model has emerged as a powerful tool for this purpose, yet significant research gaps remain. Existing studies often focus on a set of stocks without modeling stock behavior within specific sectors. Notably, there is a lack of research on the Consumer Cyclicals (CC) industry, which includes sectors such as automotive, housing, entertainment, and retail. These industries are highly sensitive to economic conditions, making them crucial for understanding broader economic impacts on stock behavior. Furthermore, we observe that many existing works neglect broader economic contexts, which is particularly important for CC analysis due to its sensitivity to economic trends. Additionally, previous studies on financial Transformer models typically use model tokens as feature vectors of multiple variates at a single timestep. This approach may not adequately capture the important relationships between dataset variates for long-term economic trends. To this end, we present the first study on financial TSF for the CC sector along with economic data. To support the research, we propose the first public benchmark dataset for the CC sector, consisting of traditional stock price time series, technical indicators, and temporal data, enriched with economic indicators. Next, we introduce an alternative tokenization approach to enhance the model’s ability to capture long-term trends by preserving information about nonlinear dependencies between dataset variates. We hypothesize that this approach helps capture long-term signals more effectively. Through a comprehensive data ablation study and benchmark testing, we demonstrate that incorporating economic indicators improves the accuracy of longer-term predictions for the CC sector, and the new tokenization method enhances the performance of Transformer models. The dataset and code are made publicly available at: https://github.com/KasperKrawczyk/econtrans_dataset
published_date	2025-10-23T05:25:17Z
_version_	1851550667505664000
score	11.090091

Financial Forecasting in Consumer Cyclicals with Economic Indicators and Tokenization

Similar Items