Using Machine Learning to Detect Misstatements

53 Pages Posted: 20 Dec 2019

See all articles by Jeremy Bertomeu

Jeremy Bertomeu

University of California, San Diego (UCSD) - Rady School of Management

Edwige Cheynel

University of California, San Diego (UCSD) - Rady School of Management

Eric Floyd

University of California San Diego

Wenqiang Pan

Columbia University - Columbia Business School

Date Written: December 1, 2019

Abstract

Machine learning offers empirical methods to sift through accounting data sets with a large number of variables and limited a priori knowledge about functional forms. In this study, we show that these methods help detect and interpret patterns present in ongoing accounting misstatements. We use a wide set of variables from accounting, capital markets, governance, and auditing datasets to detect material misstatements. A primary insight of our analysis is that accounting variables, while they do not detect misstatements well on their own, become most important with suitable interactions with audit and market variables. We also analyze differences between misstatements and irregularities, compare algorithms, examine one-year and twoyear ahead predictions, and interpret groups at greater risk of misstatements.

Keywords: Machine Learning; Big Data; Analytics; Misstatements; AAERs; Accounting Fraud

JEL Classification: C63; D83; G38; K22; K42; M41

Suggested Citation

Bertomeu, Jeremy and Cheynel, Edwige and Floyd, Eric and Pan, Wenqiang, Using Machine Learning to Detect Misstatements (December 1, 2019). Available at SSRN: https://ssrn.com/abstract=3496297

Jeremy Bertomeu (Contact Author)

University of California, San Diego (UCSD) - Rady School of Management ( email )

9500 Gilman Drive
Rady School of Management
La Jolla, CA 92093
United States

Edwige Cheynel

University of California, San Diego (UCSD) - Rady School of Management ( email )

Eric Floyd

University of California San Diego ( email )

CA
United States

Wenqiang Pan

Columbia University - Columbia Business School ( email )

New York, NY
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
165
Abstract Views
478
rank
186,075
PlumX Metrics