Auditing Section Research Summaries Space

A Database of Auditing Research - Building Bridges with Practice

This is a public Custom Hive  public

research summary

    Detecting and Predicting Accounting Irregularities: A...
    research summary posted October 20, 2014 by Jennifer M Mueller-Phillips, tagged 06.0 Risk and Risk Management, Including Fraud Risk, 06.02 Fraud Risk Models, 06.05 Assessing Risk of Material Misstatement 
    Detecting and Predicting Accounting Irregularities: A Comparison of Commercial and Academic Risk Measures
    Practical Implications:

    The results of this study should be useful to research practitioners, regulators, investors, auditors (internal and external), managers, boards of directors, and analysts.  Academic researchers who study fraud or aggressive financial reporting should also be interested in understanding which risk measures have the highest statistical power and construct validity.  One clear advantage of the academic risk measures is that, unlike commercially developed risk measures that are proprietary by nature, researchers know all of the inputs to the academic measures.  On the other hand, studies that need an overall estimate of ex ante financial reporting risk or studies with small or limited sample sizes are likely to benefit the most from using comprehensive, commercially developed risk measures like AGR due to its improved statistical power.


    For more information on this study, please contact David A. Wood.


    Price III, R. A., N. Y. Sharp, and D. A. Wood. 2011. Detecting and predicting accounting irregularities:  A comparison of commercial and academic risk measures. Accounting Horizons 25 (4): 755-780

    Accounting irregularities, detecting fraud, predicting fraud, risk measures, commercial risk ratings
    Purpose of the Study:

    A substantial body of academic research is devoted to developing and testing risk proxies that detect accounting irregularities but the academic literature has paid little attention to commercially developed risk measures.  The authors compare the commercially developed Accounting and Governance Risk (AGR) and Accounting Risk (AR) measures with academic risk measures to determine which best detects financial misstatements that result in: (1) Securities and Exchange Commission enforcement actions; (2) egregious accounting restatements; and (3) shareholder lawsuits related to accounting improprieties.  By making this comparison, the authors provide evidence concerning which metrics among the academic and commercial measures have the greatest ability to detect accounting irregularities and, second, to provide insight on the attribute of superior risk metrics. The authors of this paper also include relatively new academic risk measures that were unavailable for examination in prior studies that compared academic risk measures. 

    Design/Method/ Approach:

    The research sample data is based on the set of firms in Compustat from 1995 to 2008.  Data was obtained from several additional sources, including the Center for Research on Security Prices, Audit Analytics, and Audit Integrity.  The AGR measure is produced commercially by Audit Integrity to estimate the likelihood that reported financial information includes elements that are misleading or fraudulent.  Multivariate logistic regressions were run to test the detective and predictive ability of the measures.

    • The authors find the commercially developed risk measure AGR is useful as a proxy for the risk of accounting irregularities.
    • In comparisons between AGR and academic risk measures, the authors find that AGR performs as well as or better than academic risk measures in all comparisons.
    • The authors find that among the academic risk measures, accrual estimation errors and unexplained audit fees appear to perform the best.
    Risk & Risk Management - Including Fraud Risk
    Assessing Risk of Material Misstatement, Fraud Risk Models