What are you saying? Using topic to detect financial misreporting

We use a machine learning technique to assess whether the thematic content of financial statement disclosures (labeled topic) is incrementally informative in predicting intentional misreporting. Using a Bayesian topic modeling algorithm, we determine and empirically quantify the topic content of a l...

Full description

Saved in:
Bibliographic Details
Main Authors: BROWN, Nerissa C., CROWLEY, Richard M., ELLIOTT, W. Brooke
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2020
Subjects:
Online Access:https://ink.library.smu.edu.sg/soa_research/1828
https://ink.library.smu.edu.sg/context/soa_research/article/2855/viewcontent/BROWN_et_al_2020_Journal_of_Accounting_Research.pdf
https://ink.library.smu.edu.sg/context/soa_research/article/2855/filename/0/type/additional/viewcontent/BCE_Online_Appendix.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.soa_research-2855
record_format dspace
spelling sg-smu-ink.soa_research-28552020-02-28T02:23:26Z What are you saying? Using topic to detect financial misreporting BROWN, Nerissa C. CROWLEY, Richard M. ELLIOTT, W. Brooke We use a machine learning technique to assess whether the thematic content of financial statement disclosures (labeled topic) is incrementally informative in predicting intentional misreporting. Using a Bayesian topic modeling algorithm, we determine and empirically quantify the topic content of a large collection of 10‐K narratives spanning 1994 to 2012. We find that the algorithm produces a valid set of semantically meaningful topics that predict financial misreporting, based on samples of Securities and Exchange Commission (SEC) enforcement actions (Accounting and Auditing Enforcement Releases [AAERs]) and irregularities identified from financial restatements and 10‐K filing amendments. Our out‐of‐sample tests indicate that topic significantly improves the detection of financial misreporting by as much as 59% when added to models based on commonly used financial and textual style variables. Furthermore, models that incorporate topic significantly outperform traditional models when detecting serious revenue recognition and core expense errors. Taken together, our results suggest that the topics discussed in annual report filings and the attention devoted to each topic are useful signals in detecting financial misreporting. 2020-03-01T08:00:00Z text application/pdf https://ink.library.smu.edu.sg/soa_research/1828 info:doi/10.1111/1475-679X.12294 https://ink.library.smu.edu.sg/context/soa_research/article/2855/viewcontent/BROWN_et_al_2020_Journal_of_Accounting_Research.pdf https://ink.library.smu.edu.sg/context/soa_research/article/2855/filename/0/type/additional/viewcontent/BCE_Online_Appendix.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Accountancy eng Institutional Knowledge at Singapore Management University topic modeling disclosure latent Dirichlet allocation financial misreporting Accounting Corporate Finance
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic topic modeling
disclosure
latent Dirichlet allocation
financial misreporting
Accounting
Corporate Finance
spellingShingle topic modeling
disclosure
latent Dirichlet allocation
financial misreporting
Accounting
Corporate Finance
BROWN, Nerissa C.
CROWLEY, Richard M.
ELLIOTT, W. Brooke
What are you saying? Using topic to detect financial misreporting
description We use a machine learning technique to assess whether the thematic content of financial statement disclosures (labeled topic) is incrementally informative in predicting intentional misreporting. Using a Bayesian topic modeling algorithm, we determine and empirically quantify the topic content of a large collection of 10‐K narratives spanning 1994 to 2012. We find that the algorithm produces a valid set of semantically meaningful topics that predict financial misreporting, based on samples of Securities and Exchange Commission (SEC) enforcement actions (Accounting and Auditing Enforcement Releases [AAERs]) and irregularities identified from financial restatements and 10‐K filing amendments. Our out‐of‐sample tests indicate that topic significantly improves the detection of financial misreporting by as much as 59% when added to models based on commonly used financial and textual style variables. Furthermore, models that incorporate topic significantly outperform traditional models when detecting serious revenue recognition and core expense errors. Taken together, our results suggest that the topics discussed in annual report filings and the attention devoted to each topic are useful signals in detecting financial misreporting.
format text
author BROWN, Nerissa C.
CROWLEY, Richard M.
ELLIOTT, W. Brooke
author_facet BROWN, Nerissa C.
CROWLEY, Richard M.
ELLIOTT, W. Brooke
author_sort BROWN, Nerissa C.
title What are you saying? Using topic to detect financial misreporting
title_short What are you saying? Using topic to detect financial misreporting
title_full What are you saying? Using topic to detect financial misreporting
title_fullStr What are you saying? Using topic to detect financial misreporting
title_full_unstemmed What are you saying? Using topic to detect financial misreporting
title_sort what are you saying? using topic to detect financial misreporting
publisher Institutional Knowledge at Singapore Management University
publishDate 2020
url https://ink.library.smu.edu.sg/soa_research/1828
https://ink.library.smu.edu.sg/context/soa_research/article/2855/viewcontent/BROWN_et_al_2020_Journal_of_Accounting_Research.pdf
https://ink.library.smu.edu.sg/context/soa_research/article/2855/filename/0/type/additional/viewcontent/BCE_Online_Appendix.pdf
_version_ 1770575102003380224