An International Meeting of the American Accounting Association American Accounting Association 2006 Annual Meeting August 6–9, 2006 Washington, D.C.

Benjamin P. Foster
University of Louisville

Jozef Zurada
University of Louisville

Terry J. Ward
Middle Tennessee State University

Abstract: A sometimes critical and difficult decision faced by auditors is whether to modify their audit opinion based on the going-concern assumption. Auditors face economic tradeoffs in their decision whether to issue a going-concern opinion. This study examines whether data mining techniques including decision trees can assist auditors in their going-concern decisions. We evaluated the predictive ability of our statistical methods based on information auditors would have before issuing an audit opinion on companies' 1999 financial statements. A relatively simple decision tree provides the best overall classification accuracy. Consequently, decision trees could perhaps assist auditors in their going-concern decisions and provide an additional defense against litigation.

Back to Session Listing

AAA Home Page