No Cover Image

Journal article 423 views 159 downloads

Tax Default Prediction Using Feature Transformation-Based Machine Learning

Mohammad Abedin Orcid Logo, Guotai Chi Orcid Logo, Mohammed Mohi Uddin, Md. Shahriare Satu, Md. Imran Khan Orcid Logo, Petr Hajek Orcid Logo

IEEE Access, Volume: 9, Pages: 19864 - 19881

Swansea University Author: Mohammad Abedin Orcid Logo

  • 64271.VOR.pdf

    PDF | Version of Record

    © Author(s) 2020. Distributed under the terms of a Creative Commons Attribution 4.0 License (CC BY 4.0).

    Download (4.03MB)

Check full text

DOI (Published version): 10.1109/access.2020.3048018

Abstract

This study proposes to address the economic significance of unpaid taxes by using an automatic system for predicting a tax default. Too little attention has been paid to tax default prediction in the past. Moreover, existing approaches tend to apply conventional statistical methods rather than advan...

Full description

Published in: IEEE Access
ISSN: 2169-3536
Published: Institute of Electrical and Electronics Engineers (IEEE) 2021
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa64271
first_indexed 2023-09-18T13:46:41Z
last_indexed 2024-11-25T14:13:45Z
id cronfa64271
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2023-09-19T16:14:23.2796578</datestamp><bib-version>v2</bib-version><id>64271</id><entry>2023-08-31</entry><title>Tax Default Prediction Using Feature Transformation-Based Machine Learning</title><swanseaauthors><author><sid>4ed8c020eae0c9bec4f5d9495d86d415</sid><ORCID>0000-0002-4688-0619</ORCID><firstname>Mohammad</firstname><surname>Abedin</surname><name>Mohammad Abedin</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2023-08-31</date><deptcode>CBAE</deptcode><abstract>This study proposes to address the economic significance of unpaid taxes by using an automatic system for predicting a tax default. Too little attention has been paid to tax default prediction in the past. Moreover, existing approaches tend to apply conventional statistical methods rather than advanced data analytic approaches, including state-of-the-art machine learning methods. Therefore, existing studies cannot effectively detect tax default information in real-world financial data because they fail to take into account the appropriate data transformations and nonlinear relationships between early-warning financial indicators and tax default behavior. To overcome these problems, this study applies diverse feature transformation techniques and state-of-the-art machine learning approaches. The proposed prediction system is validated by using a dataset showing tax defaults and non-defaults at Finnish limited liability firms. Our findings provide evidence for a major role of feature transformation, such as logarithmic and square-root transformation, in improving the performance of tax default prediction. We also show that extreme gradient boosting and the systematically developed forest of multiple decision trees outperform other machine learning methods in terms of accuracy and other classification performance measures. We show that the equity ratio, liquidity ratio, and debt-to-sales ratio are the most important indicators of tax defaults for 1-year-ahead predictions. Therefore, this study highlights the essential role of well-designed tax default prediction systems, which require a combination of feature transformation and machine learning methods. The effective implementation of an automatic tax default prediction system has important implications for tax administration and can assist administrators in achieving feasible government expenditure allocations and revenue expansions.</abstract><type>Journal Article</type><journal>IEEE Access</journal><volume>9</volume><journalNumber/><paginationStart>19864</paginationStart><paginationEnd>19881</paginationEnd><publisher>Institute of Electrical and Electronics Engineers (IEEE)</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic>2169-3536</issnElectronic><keywords>Default prediction, corporate tax, machine learning, feature transformation.</keywords><publishedDay>3</publishedDay><publishedMonth>2</publishedMonth><publishedYear>2021</publishedYear><publishedDate>2021-02-03</publishedDate><doi>10.1109/access.2020.3048018</doi><url>http://dx.doi.org/10.1109/access.2020.3048018</url><notes/><college>COLLEGE NANME</college><department>Management School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>CBAE</DepartmentCode><institution>Swansea University</institution><apcterm/><funders>This work has been supported by the Key Projects of National Natural Science Foundation of China (71731003), the General Projects of National Natural Science Foundation of China (72071026, 71873103, 71971051, and 71971034), the Youth Projects of National Natural Science Foundation of China (71901055, and 71903019), the Major Projects of National Social Science Foundation of China (18ZDA095), the scientific research project of the Czech Sciences Foundation Grant No. 19-15498S. This paper has also been supported by the Bank of Dalian and Postal Savings Bank of China. We thank the organizations mentioned above.</funders><projectreference/><lastEdited>2023-09-19T16:14:23.2796578</lastEdited><Created>2023-08-31T19:05:51.6971175</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>Mohammad</firstname><surname>Abedin</surname><orcid>0000-0002-4688-0619</orcid><order>1</order></author><author><firstname>Guotai</firstname><surname>Chi</surname><orcid>0000-0002-4975-4394</orcid><order>2</order></author><author><firstname>Mohammed Mohi</firstname><surname>Uddin</surname><order>3</order></author><author><firstname>Md. Shahriare</firstname><surname>Satu</surname><order>4</order></author><author><firstname>Md. Imran</firstname><surname>Khan</surname><orcid>0000-0002-7398-5280</orcid><order>5</order></author><author><firstname>Petr</firstname><surname>Hajek</surname><orcid>0000-0001-5579-1215</orcid><order>6</order></author></authors><documents><document><filename>64271__28559__f98ebeadc4d84a69854da663ca1ed08b.pdf</filename><originalFilename>64271.VOR.pdf</originalFilename><uploaded>2023-09-18T14:43:19.7115763</uploaded><type>Output</type><contentLength>4227664</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>&#xA9; Author(s) 2020. Distributed under the terms of a Creative Commons Attribution 4.0 License (CC BY 4.0).</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2023-09-19T16:14:23.2796578 v2 64271 2023-08-31 Tax Default Prediction Using Feature Transformation-Based Machine Learning 4ed8c020eae0c9bec4f5d9495d86d415 0000-0002-4688-0619 Mohammad Abedin Mohammad Abedin true false 2023-08-31 CBAE This study proposes to address the economic significance of unpaid taxes by using an automatic system for predicting a tax default. Too little attention has been paid to tax default prediction in the past. Moreover, existing approaches tend to apply conventional statistical methods rather than advanced data analytic approaches, including state-of-the-art machine learning methods. Therefore, existing studies cannot effectively detect tax default information in real-world financial data because they fail to take into account the appropriate data transformations and nonlinear relationships between early-warning financial indicators and tax default behavior. To overcome these problems, this study applies diverse feature transformation techniques and state-of-the-art machine learning approaches. The proposed prediction system is validated by using a dataset showing tax defaults and non-defaults at Finnish limited liability firms. Our findings provide evidence for a major role of feature transformation, such as logarithmic and square-root transformation, in improving the performance of tax default prediction. We also show that extreme gradient boosting and the systematically developed forest of multiple decision trees outperform other machine learning methods in terms of accuracy and other classification performance measures. We show that the equity ratio, liquidity ratio, and debt-to-sales ratio are the most important indicators of tax defaults for 1-year-ahead predictions. Therefore, this study highlights the essential role of well-designed tax default prediction systems, which require a combination of feature transformation and machine learning methods. The effective implementation of an automatic tax default prediction system has important implications for tax administration and can assist administrators in achieving feasible government expenditure allocations and revenue expansions. Journal Article IEEE Access 9 19864 19881 Institute of Electrical and Electronics Engineers (IEEE) 2169-3536 Default prediction, corporate tax, machine learning, feature transformation. 3 2 2021 2021-02-03 10.1109/access.2020.3048018 http://dx.doi.org/10.1109/access.2020.3048018 COLLEGE NANME Management School COLLEGE CODE CBAE Swansea University This work has been supported by the Key Projects of National Natural Science Foundation of China (71731003), the General Projects of National Natural Science Foundation of China (72071026, 71873103, 71971051, and 71971034), the Youth Projects of National Natural Science Foundation of China (71901055, and 71903019), the Major Projects of National Social Science Foundation of China (18ZDA095), the scientific research project of the Czech Sciences Foundation Grant No. 19-15498S. This paper has also been supported by the Bank of Dalian and Postal Savings Bank of China. We thank the organizations mentioned above. 2023-09-19T16:14:23.2796578 2023-08-31T19:05:51.6971175 Faculty of Humanities and Social Sciences School of Management - Accounting and Finance Mohammad Abedin 0000-0002-4688-0619 1 Guotai Chi 0000-0002-4975-4394 2 Mohammed Mohi Uddin 3 Md. Shahriare Satu 4 Md. Imran Khan 0000-0002-7398-5280 5 Petr Hajek 0000-0001-5579-1215 6 64271__28559__f98ebeadc4d84a69854da663ca1ed08b.pdf 64271.VOR.pdf 2023-09-18T14:43:19.7115763 Output 4227664 application/pdf Version of Record true © Author(s) 2020. Distributed under the terms of a Creative Commons Attribution 4.0 License (CC BY 4.0). true eng https://creativecommons.org/licenses/by/4.0/
title Tax Default Prediction Using Feature Transformation-Based Machine Learning
spellingShingle Tax Default Prediction Using Feature Transformation-Based Machine Learning
Mohammad Abedin
title_short Tax Default Prediction Using Feature Transformation-Based Machine Learning
title_full Tax Default Prediction Using Feature Transformation-Based Machine Learning
title_fullStr Tax Default Prediction Using Feature Transformation-Based Machine Learning
title_full_unstemmed Tax Default Prediction Using Feature Transformation-Based Machine Learning
title_sort Tax Default Prediction Using Feature Transformation-Based Machine Learning
author_id_str_mv 4ed8c020eae0c9bec4f5d9495d86d415
author_id_fullname_str_mv 4ed8c020eae0c9bec4f5d9495d86d415_***_Mohammad Abedin
author Mohammad Abedin
author2 Mohammad Abedin
Guotai Chi
Mohammed Mohi Uddin
Md. Shahriare Satu
Md. Imran Khan
Petr Hajek
format Journal article
container_title IEEE Access
container_volume 9
container_start_page 19864
publishDate 2021
institution Swansea University
issn 2169-3536
doi_str_mv 10.1109/access.2020.3048018
publisher Institute of Electrical and Electronics Engineers (IEEE)
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/10.1109/access.2020.3048018
document_store_str 1
active_str 0
description This study proposes to address the economic significance of unpaid taxes by using an automatic system for predicting a tax default. Too little attention has been paid to tax default prediction in the past. Moreover, existing approaches tend to apply conventional statistical methods rather than advanced data analytic approaches, including state-of-the-art machine learning methods. Therefore, existing studies cannot effectively detect tax default information in real-world financial data because they fail to take into account the appropriate data transformations and nonlinear relationships between early-warning financial indicators and tax default behavior. To overcome these problems, this study applies diverse feature transformation techniques and state-of-the-art machine learning approaches. The proposed prediction system is validated by using a dataset showing tax defaults and non-defaults at Finnish limited liability firms. Our findings provide evidence for a major role of feature transformation, such as logarithmic and square-root transformation, in improving the performance of tax default prediction. We also show that extreme gradient boosting and the systematically developed forest of multiple decision trees outperform other machine learning methods in terms of accuracy and other classification performance measures. We show that the equity ratio, liquidity ratio, and debt-to-sales ratio are the most important indicators of tax defaults for 1-year-ahead predictions. Therefore, this study highlights the essential role of well-designed tax default prediction systems, which require a combination of feature transformation and machine learning methods. The effective implementation of an automatic tax default prediction system has important implications for tax administration and can assist administrators in achieving feasible government expenditure allocations and revenue expansions.
published_date 2021-02-03T14:27:11Z
_version_ 1821325374602084352
score 11.048042

Similar Items