Improved models to detect fraud in financial statements

Adrian Gepp, Kuldeep Kumar, Sukanto Bhattacharya

Research output: Contribution to conferenceAbstractResearchpeer-review

Abstract

Studies have estimated the median loss from a single financial statement fraud scheme to be at least one million US dollars and the annual cost of financial statement fraud could exceed 1.2 trillion US dollars worldwide. Many business decisions rely on the accuracy of financial statements, but resources are not available to comprehensively investigate all of them.

Detection of this type of fraud is difficult. Consequently, there is a need for better decision aids such as those developed in this research.

Standard parametric regression-based techniques, particularly logistic regression, have been extensively studied for detecting financial statement fraud. More investigation is needed into non-parametric techniques such as decision trees and ensemble techniques that combine multiple models. Consequently, 34 different models have been compared over a range of ratios of the cost of failing to detect fraud relative to the cost of falsely alleging it, as these costs differ among stakeholders. Some models are the same as those used in prior studies, some are modifications of previously used models, and entirely new ones have also been developed. A large number of potential explanatory variables are also investigated in order to study which are the most useful to detection models. Empirical support has been found for both financial and non-financial explanatory variables, including new variables.

New models developed in this research outperform extant ones on holdout data. Using these models, financial statements can be automatically classified as either fraudulent or legitimate, as well as be ranked according to their likelihood of being fraudulent. This information can then be used to improve early detection, which would mitigate fraud’s cost and help deter its future occurrence.
Original languageEnglish
Publication statusPublished - 2016
Event28th Asian-Pacific Conference on International Accounting Issues - Maui, Hawaii, United States
Duration: 6 Nov 20169 Nov 2016
Conference number: 28
http://www.apconference.org/28th_APC/

Conference

Conference28th Asian-Pacific Conference on International Accounting Issues
CountryUnited States
CityMaui, Hawaii
Period6/11/169/11/16
Internet address

Fingerprint

Financial statements
Fraud
Costs
Financial statement fraud
Median
Resources
Decision tree
Decision aids
Logistic regression
Stakeholders

Cite this

Gepp, A., Kumar, K., & Bhattacharya, S. (2016). Improved models to detect fraud in financial statements. Abstract from 28th Asian-Pacific Conference on International Accounting Issues, Maui, Hawaii, United States.
Gepp, Adrian ; Kumar, Kuldeep ; Bhattacharya, Sukanto. / Improved models to detect fraud in financial statements. Abstract from 28th Asian-Pacific Conference on International Accounting Issues, Maui, Hawaii, United States.
@conference{ae36cd4e2d1b4fe8a16b4185a6e46bf2,
title = "Improved models to detect fraud in financial statements",
abstract = "Studies have estimated the median loss from a single financial statement fraud scheme to be at least one million US dollars and the annual cost of financial statement fraud could exceed 1.2 trillion US dollars worldwide. Many business decisions rely on the accuracy of financial statements, but resources are not available to comprehensively investigate all of them.Detection of this type of fraud is difficult. Consequently, there is a need for better decision aids such as those developed in this research.Standard parametric regression-based techniques, particularly logistic regression, have been extensively studied for detecting financial statement fraud. More investigation is needed into non-parametric techniques such as decision trees and ensemble techniques that combine multiple models. Consequently, 34 different models have been compared over a range of ratios of the cost of failing to detect fraud relative to the cost of falsely alleging it, as these costs differ among stakeholders. Some models are the same as those used in prior studies, some are modifications of previously used models, and entirely new ones have also been developed. A large number of potential explanatory variables are also investigated in order to study which are the most useful to detection models. Empirical support has been found for both financial and non-financial explanatory variables, including new variables.New models developed in this research outperform extant ones on holdout data. Using these models, financial statements can be automatically classified as either fraudulent or legitimate, as well as be ranked according to their likelihood of being fraudulent. This information can then be used to improve early detection, which would mitigate fraud’s cost and help deter its future occurrence.",
author = "Adrian Gepp and Kuldeep Kumar and Sukanto Bhattacharya",
year = "2016",
language = "English",
note = "28th Asian-Pacific Conference on International Accounting Issues ; Conference date: 06-11-2016 Through 09-11-2016",
url = "http://www.apconference.org/28th_APC/",

}

Gepp, A, Kumar, K & Bhattacharya, S 2016, 'Improved models to detect fraud in financial statements' 28th Asian-Pacific Conference on International Accounting Issues, Maui, Hawaii, United States, 6/11/16 - 9/11/16, .

Improved models to detect fraud in financial statements. / Gepp, Adrian; Kumar, Kuldeep; Bhattacharya, Sukanto.

2016. Abstract from 28th Asian-Pacific Conference on International Accounting Issues, Maui, Hawaii, United States.

Research output: Contribution to conferenceAbstractResearchpeer-review

TY - CONF

T1 - Improved models to detect fraud in financial statements

AU - Gepp, Adrian

AU - Kumar, Kuldeep

AU - Bhattacharya, Sukanto

PY - 2016

Y1 - 2016

N2 - Studies have estimated the median loss from a single financial statement fraud scheme to be at least one million US dollars and the annual cost of financial statement fraud could exceed 1.2 trillion US dollars worldwide. Many business decisions rely on the accuracy of financial statements, but resources are not available to comprehensively investigate all of them.Detection of this type of fraud is difficult. Consequently, there is a need for better decision aids such as those developed in this research.Standard parametric regression-based techniques, particularly logistic regression, have been extensively studied for detecting financial statement fraud. More investigation is needed into non-parametric techniques such as decision trees and ensemble techniques that combine multiple models. Consequently, 34 different models have been compared over a range of ratios of the cost of failing to detect fraud relative to the cost of falsely alleging it, as these costs differ among stakeholders. Some models are the same as those used in prior studies, some are modifications of previously used models, and entirely new ones have also been developed. A large number of potential explanatory variables are also investigated in order to study which are the most useful to detection models. Empirical support has been found for both financial and non-financial explanatory variables, including new variables.New models developed in this research outperform extant ones on holdout data. Using these models, financial statements can be automatically classified as either fraudulent or legitimate, as well as be ranked according to their likelihood of being fraudulent. This information can then be used to improve early detection, which would mitigate fraud’s cost and help deter its future occurrence.

AB - Studies have estimated the median loss from a single financial statement fraud scheme to be at least one million US dollars and the annual cost of financial statement fraud could exceed 1.2 trillion US dollars worldwide. Many business decisions rely on the accuracy of financial statements, but resources are not available to comprehensively investigate all of them.Detection of this type of fraud is difficult. Consequently, there is a need for better decision aids such as those developed in this research.Standard parametric regression-based techniques, particularly logistic regression, have been extensively studied for detecting financial statement fraud. More investigation is needed into non-parametric techniques such as decision trees and ensemble techniques that combine multiple models. Consequently, 34 different models have been compared over a range of ratios of the cost of failing to detect fraud relative to the cost of falsely alleging it, as these costs differ among stakeholders. Some models are the same as those used in prior studies, some are modifications of previously used models, and entirely new ones have also been developed. A large number of potential explanatory variables are also investigated in order to study which are the most useful to detection models. Empirical support has been found for both financial and non-financial explanatory variables, including new variables.New models developed in this research outperform extant ones on holdout data. Using these models, financial statements can be automatically classified as either fraudulent or legitimate, as well as be ranked according to their likelihood of being fraudulent. This information can then be used to improve early detection, which would mitigate fraud’s cost and help deter its future occurrence.

UR - http://www.apconference.org/28th_APC/documents/28th_APC.pdf

M3 - Abstract

ER -

Gepp A, Kumar K, Bhattacharya S. Improved models to detect fraud in financial statements. 2016. Abstract from 28th Asian-Pacific Conference on International Accounting Issues, Maui, Hawaii, United States.