No Cover Image

Journal article 151 views

Deep-learning model using hybrid adaptive trend estimated series for modelling and forecasting sales

Md. Iftekharul Alam Efat, Petr Hajek Orcid Logo, Abedin Abedin, Rahat Uddin Azad, Md. Al Jaber, Shuvra Aditya, Mohammad Kabir Hassan

Annals of Operations Research

Swansea University Author: Abedin Abedin

Full text not available from this repository: check for access using links below.

Check full text

DOI (Published version): 10.1007/s10479-022-04838-6

Abstract

Existing sales forecasting models are not comprehensive and flexible enough to consider dynamic changes and nonlinearities in sales time-series at the store and product levels. To capture different big data characteristics in sales forecasting data, such as seasonal and trend variations, this study...

Full description

Published in: Annals of Operations Research
ISSN: 0254-5330 1572-9338
Published: Springer Science and Business Media LLC
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa64231
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2023-09-20T14:17:53Z
last_indexed 2023-09-20T14:17:53Z
id cronfa64231
recordtype SURis
fullrecord <?xml version="1.0" encoding="utf-8"?><rfc1807 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema"><bib-version>v2</bib-version><id>64231</id><entry>2023-08-31</entry><title>Deep-learning model using hybrid adaptive trend estimated series for modelling and forecasting sales</title><swanseaauthors><author><sid>4ed8c020eae0c9bec4f5d9495d86d415</sid><firstname>Abedin</firstname><surname>Abedin</surname><name>Abedin Abedin</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2023-08-31</date><deptcode>BAF</deptcode><abstract>Existing sales forecasting models are not comprehensive and flexible enough to consider dynamic changes and nonlinearities in sales time-series at the store and product levels. To capture different big data characteristics in sales forecasting data, such as seasonal and trend variations, this study develops a hybrid model combining adaptive trend estimated series (ATES) with a deep neural network model. ATES is first used to model seasonal effects and incorporate holiday, weekend, and marketing effects on sales. The deep neural network model is then proposed to model residuals by capturing complex high-level spatiotemporal features from the data. The proposed hybrid model is equipped with a feature-extraction component that automatically detects the patterns and trends in time-series, which makes the forecasting model robust against noise and time-series length. To validate the proposed hybrid model, a large volume of sales data is processed with a three-dimensional data model to effectively support business decisions at the product-specific store level. To demonstrate the effectiveness of the proposed model, a comparative analysis is performed with several state-of-the-art sales forecasting methods. Here, we show that the proposed hybrid model outperforms existing models for forecasting horizons ranging from one to 12 months.</abstract><type>Journal Article</type><journal>Annals of Operations Research</journal><volume/><journalNumber/><paginationStart/><paginationEnd/><publisher>Springer Science and Business Media LLC</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>0254-5330</issnPrint><issnElectronic>1572-9338</issnElectronic><keywords>Machine learning, Sales forecasting, Big data, Regression model, Deep learning</keywords><publishedDay>0</publishedDay><publishedMonth>0</publishedMonth><publishedYear>0</publishedYear><publishedDate>0001-01-01</publishedDate><doi>10.1007/s10479-022-04838-6</doi><url>http://dx.doi.org/10.1007/s10479-022-04838-6</url><notes/><college>COLLEGE NANME</college><department>Accounting and Finance</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>BAF</DepartmentCode><institution>Swansea University</institution><apcterm/><funders>This article was directed by Software Evaluation and Re-Engineering Research Lab (SERER Lab) and supported by the scientific research project of the Czech Sciences Foundation Grant No. 19-15498S.</funders><projectreference/><lastEdited>2023-09-22T15:19:53.2932562</lastEdited><Created>2023-08-31T17:33:27.1122645</Created><path><level id="1">Faculty of Humanities and Social Sciences</level><level id="2">School of Management - Accounting and Finance</level></path><authors><author><firstname>Md. Iftekharul Alam</firstname><surname>Efat</surname><order>1</order></author><author><firstname>Petr</firstname><surname>Hajek</surname><orcid>0000-0001-5579-1215</orcid><order>2</order></author><author><firstname>Abedin</firstname><surname>Abedin</surname><order>3</order></author><author><firstname>Rahat Uddin</firstname><surname>Azad</surname><order>4</order></author><author><firstname>Md. Al</firstname><surname>Jaber</surname><order>5</order></author><author><firstname>Shuvra</firstname><surname>Aditya</surname><order>6</order></author><author><firstname>Mohammad Kabir</firstname><surname>Hassan</surname><order>7</order></author></authors><documents/><OutputDurs/></rfc1807>
spelling v2 64231 2023-08-31 Deep-learning model using hybrid adaptive trend estimated series for modelling and forecasting sales 4ed8c020eae0c9bec4f5d9495d86d415 Abedin Abedin Abedin Abedin true false 2023-08-31 BAF Existing sales forecasting models are not comprehensive and flexible enough to consider dynamic changes and nonlinearities in sales time-series at the store and product levels. To capture different big data characteristics in sales forecasting data, such as seasonal and trend variations, this study develops a hybrid model combining adaptive trend estimated series (ATES) with a deep neural network model. ATES is first used to model seasonal effects and incorporate holiday, weekend, and marketing effects on sales. The deep neural network model is then proposed to model residuals by capturing complex high-level spatiotemporal features from the data. The proposed hybrid model is equipped with a feature-extraction component that automatically detects the patterns and trends in time-series, which makes the forecasting model robust against noise and time-series length. To validate the proposed hybrid model, a large volume of sales data is processed with a three-dimensional data model to effectively support business decisions at the product-specific store level. To demonstrate the effectiveness of the proposed model, a comparative analysis is performed with several state-of-the-art sales forecasting methods. Here, we show that the proposed hybrid model outperforms existing models for forecasting horizons ranging from one to 12 months. Journal Article Annals of Operations Research Springer Science and Business Media LLC 0254-5330 1572-9338 Machine learning, Sales forecasting, Big data, Regression model, Deep learning 0 0 0 0001-01-01 10.1007/s10479-022-04838-6 http://dx.doi.org/10.1007/s10479-022-04838-6 COLLEGE NANME Accounting and Finance COLLEGE CODE BAF Swansea University This article was directed by Software Evaluation and Re-Engineering Research Lab (SERER Lab) and supported by the scientific research project of the Czech Sciences Foundation Grant No. 19-15498S. 2023-09-22T15:19:53.2932562 2023-08-31T17:33:27.1122645 Faculty of Humanities and Social Sciences School of Management - Accounting and Finance Md. Iftekharul Alam Efat 1 Petr Hajek 0000-0001-5579-1215 2 Abedin Abedin 3 Rahat Uddin Azad 4 Md. Al Jaber 5 Shuvra Aditya 6 Mohammad Kabir Hassan 7
title Deep-learning model using hybrid adaptive trend estimated series for modelling and forecasting sales
spellingShingle Deep-learning model using hybrid adaptive trend estimated series for modelling and forecasting sales
Abedin Abedin
title_short Deep-learning model using hybrid adaptive trend estimated series for modelling and forecasting sales
title_full Deep-learning model using hybrid adaptive trend estimated series for modelling and forecasting sales
title_fullStr Deep-learning model using hybrid adaptive trend estimated series for modelling and forecasting sales
title_full_unstemmed Deep-learning model using hybrid adaptive trend estimated series for modelling and forecasting sales
title_sort Deep-learning model using hybrid adaptive trend estimated series for modelling and forecasting sales
author_id_str_mv 4ed8c020eae0c9bec4f5d9495d86d415
author_id_fullname_str_mv 4ed8c020eae0c9bec4f5d9495d86d415_***_Abedin Abedin
author Abedin Abedin
author2 Md. Iftekharul Alam Efat
Petr Hajek
Abedin Abedin
Rahat Uddin Azad
Md. Al Jaber
Shuvra Aditya
Mohammad Kabir Hassan
format Journal article
container_title Annals of Operations Research
institution Swansea University
issn 0254-5330
1572-9338
doi_str_mv 10.1007/s10479-022-04838-6
publisher Springer Science and Business Media LLC
college_str Faculty of Humanities and Social Sciences
hierarchytype
hierarchy_top_id facultyofhumanitiesandsocialsciences
hierarchy_top_title Faculty of Humanities and Social Sciences
hierarchy_parent_id facultyofhumanitiesandsocialsciences
hierarchy_parent_title Faculty of Humanities and Social Sciences
department_str School of Management - Accounting and Finance{{{_:::_}}}Faculty of Humanities and Social Sciences{{{_:::_}}}School of Management - Accounting and Finance
url http://dx.doi.org/10.1007/s10479-022-04838-6
document_store_str 0
active_str 0
description Existing sales forecasting models are not comprehensive and flexible enough to consider dynamic changes and nonlinearities in sales time-series at the store and product levels. To capture different big data characteristics in sales forecasting data, such as seasonal and trend variations, this study develops a hybrid model combining adaptive trend estimated series (ATES) with a deep neural network model. ATES is first used to model seasonal effects and incorporate holiday, weekend, and marketing effects on sales. The deep neural network model is then proposed to model residuals by capturing complex high-level spatiotemporal features from the data. The proposed hybrid model is equipped with a feature-extraction component that automatically detects the patterns and trends in time-series, which makes the forecasting model robust against noise and time-series length. To validate the proposed hybrid model, a large volume of sales data is processed with a three-dimensional data model to effectively support business decisions at the product-specific store level. To demonstrate the effectiveness of the proposed model, a comparative analysis is performed with several state-of-the-art sales forecasting methods. Here, we show that the proposed hybrid model outperforms existing models for forecasting horizons ranging from one to 12 months.
published_date 0001-01-01T15:19:52Z
_version_ 1777747773130539008
score 11.012678

Similar Items