A comparison of nearest neighbours, discriminant and logit models for auditing decisions

This study investigates the efficiency of k-Nearest Neighbours (k-NN) in developing models for estimating auditors' opinion, as opposed to models developed with discriminant and logit analyses. The sample consists of 5,276 financial statements, out of which 980 received a qualified audit opinion, obtained from 1,455 private and public UK companies operating in the manufacturing and trade sectors. We develop two industry-specific models and a general one using data from the period 1998-2001, which are then tested over the period 2002-2003. In each case, two versions of the models are developed. The first includes only financial variables. The second includes both financial and non-financial variables. The results indicate that the inclusion of credit rating in the models results in a considerable increase both in terms of goodness of fit and classification accuracies. The comparison of the methods reveals that the k-NN models can be more efficient, in terms of average classification accuracy, than the discriminant and logit models. Finally, the results are mixed as it concerns the development of industry-specific models as opposed to general ones.

[1]  Stella Fearnley,et al.  The regulatory framework for financial reporting and auditing in the United Kingdom: the present position and impending changes☆ , 2003 .

[2]  Jayashree Mahajan.  The Overconfidence Effect in Marketing Management Predictions , 1992 .

[3]  Carol E. Brown,et al.  Opportunities for artificial intelligence development in the accounting domain: the case for auditing , 2006, Intell. Syst. Account. Finance Manag..

[4]  Ivan Bratko,et al.  VizRank: Data Visualization Guided by Machine Learning , 2006, Data Mining and Knowledge Discovery.

[5]  Roger M. Stein.  Benchmarking default prediction models: pitfalls and remedies in model validation , 2007 .

[6]  Charalambos Spathis,et al.  Audit Qualification, Firm Litigation, and Financial Information: an Empirical Analysis in Greece , 2003 .

[7]  J. Kenneth Reynolds,et al.  Does size matter? The influence of large clients on office-level auditor reporting decisions , 2000 .

[8]  Krishna G. Palepu,et al.  Predicting takeover targets: A methodological and empirical analysis , 1986 .

[9]  Keinosuke Fukunaga,et al.  An Optimal Global Nearest Neighbor Metric , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[10]  Gregory R. Madey,et al.  The Application of Neural Networks and a Qualitative Response Model to the Auditor's Going Concern Uncertainty Decision* , 1995 .

[11]  Stuart Oskamp.  Judgment under uncertainty: Overconfidence in case-study judgments , 1982 .

[12]  J. Sweeney,et al.  Fraudulently Misstated Financial Statements and Insider Trading: An Empirical Analysis , 1997 .

[13]  C. Zopounidis,et al.  Detecting falsified financial statements: a comparative study using multicriteria analysis and multivariate statistical techniques , 2002 .

[14]  Ignace De Beelde,et al.  An exploratory investigation of industry specialization of large audit firms , 1997 .

[15]  Michael Doumpos,et al.  Explaining qualifications in audit reports using a support vector machine methodology , 2005, Intell. Syst. Account. Finance Manag..

[16]  K. Ben Nowman,et al.  Continuous time and nonparametric modelling of U.S. interest rate models , 2003 .

[17]  Melody Y. Kiang,et al.  Managerial Applications of Neural Networks: The Case of Bank Failure Predictions , 1992 .

[18]  Robert Libby,et al.  Expertise And Auditors Judgments Of Conjunctive Events , 1986 .

[19]  S. Oskamp.  OVERCONFIDENCE IN CASE-STUDY JUDGMENTS. , 1965, Journal of consulting psychology.

[20]  Kenneth B. Schwartz,et al.  An empirical investigation of audit qualification decisions in the presence of going concern uncertainties , 1987 .

[21]  Michelle M. Hamer.  Failure prediction: Sensitivity of classification accuracy to alternative statistical methods and variable sets , 1983 .

[22]  Roger H. Hermanson,et al.  Audit Quality Attributes: The Perceptions of Partners, Preparers, and Financial Statement Users , 1992 .

[23]  J. Dietrich.  Discussion of Methodological Issues Related to the Estimation of Financial Distress Prediction Models , 1984 .

[24]  Paul Barnes,et al.  THE PREDICTION OF TAKEOVER TARGETS IN THE U.K. BY MEANS OF MULTIPLE DISCRIMINANT ANALYSIS , 1990 .

[25]  Pooran Wynarczyk,et al.  The Small Company Audit Qualification: A Preliminary Investigation , 1988 .

[26]  Rajendra P. Srivastava,et al.  Detection of Management Fraud: A Neural Network Approach , 1995 .

[27]  Thomas Hellström,et al.  The relevance of trends for predictions of stock returns , 2000 .

[28]  Michael Doumpos,et al.  A Stacked generalization framework for credit risk assessment , 2002, Oper. Res..

[29]  David West,et al.  Neural network credit scoring models , 2000, Comput. Oper. Res..

[30]  Yannis Manolopoulos,et al.  Data Mining techniques for the detection of fraudulent financial statements , 2007, Expert Syst. Appl..

[31]  Baruch Fischhoff,et al.  Judgment under uncertainty: For those condemned to study the past: Heuristics and biases in hindsight , 1982 .

[32]  Constantin Zopounidis,et al.  Multicriteria decision support methodologies for auditing decisions: The case of qualified audit reports in the UK , 2007, Eur. J. Oper. Res..

[33]  T. B. O'keefe,et al.  The Production of Audit Services: Evidence from a Major Public Accounting Firm , 1994 .

[34]  Constantin Zopounidis,et al.  Using client performance measures to identify pre-engagement factors associated with qualified audit reports in Greece , 2003 .

[35]  D. Hand,et al.  A k-nearest-neighbour classifier for assessing consumer credit risk , 1996 .

[36]  Constantin Zopounidis,et al.  A stacked generalization framework for the prediction of corporate acquisitions , 2003 .

[37]  Michael Doumpos,et al.  Probabilistic neural networks for the identification of qualified audit opinions , 2007, Expert Syst. Appl..

[38]  Erkki K. Laitinen,et al.  Qualified audit reports in Finland: evidence from large companies , 1998 .

[39]  Dean A. Shepherd,et al.  Vcs' Decision Processes: Evidence Suggesting More Experience May Not Always Be Better , 2003 .

[40]  Kurt Fanning,et al.  A Comparative Analysis of Artificial Neural Networks Using Financial Distress Prediction , 1994 .

[41]  Kurt Fanning,et al.  Neural Network Detection of Management Fraud Using Published Financial Data , 1998 .

[42]  Charalambos Spathis.  Detecting false financial statements using published data: some evidence from Greece , 2002 .

[43]  Constantin Zopounidis,et al.  Application of Quantitative Techniques for the Prediction of Bank Acquisition Targets , 2005, Series on Computers and Operations Research.

[44]  Fotios Pasiouras,et al.  A Comparison and Integration of Classification Techniques for the Prediction of Small UK Firms Failure , 2006 .

[45]  Terry Caelli,et al.  Compendium of Machine Learning I: A Symbolic Learning , 1996 .

[46]  Sebastian Fritz,et al.  Restructuring the credit process: behaviour scoring for german corporates , 2000 .

[47]  Hassan Espahbodi,et al.  Binary choice models and corporate takeover , 2003 .

[48]  Jen C. Ireland,et al.  An Empirical Investigation of Determinants of Audit Reports in the UK , 2003 .

[49]  Ingoo Han,et al.  Integration of Case-based Forecasting, Neural Network, and Discriminant Analysis for Bankruptcy Prediction , 1996 .

[50]  Robert A. Eisenbeis,et al.  PITFALLS IN THE APPLICATION OF DISCRIMINANT ANALYSIS IN BUSINESS, FINANCE, AND ECONOMICS , 1977 .

[51]  Chih-Chin Lai,et al.  An empirical performance comparison of machine learning methods for spam e-mail categorization , 2004, Fourth International Conference on Hybrid Intelligent Systems (HIS'04).

[52]  Michael Doumpos,et al.  Developing and Testing Models for Replicating Credit Ratings: A Multicriteria Approach , 2005 .

[53]  Terry Caelli,et al.  A compendium of machine learning: volume 1: symbolic machine learning , 1996 .

[54]  Paul Barnes,et al.  The identification of U.K. takeover targets using published historical cost accounting data Some empirical evidence comparing logit with linear discriminant analysis and raw financial ratios with industry-relative ratios , 2000 .

[55]  Jane F. Mutchler,et al.  A MULTIVARIATE-ANALYSIS OF THE AUDITORS GOING-CONCERN OPINION DECISION , 1985 .

[56]  Clive S. Lennox,et al.  Do companies successfully engage in opinion-shopping? Evidence from the UK , 2000 .

[57]  Barry L. Lewis,et al.  Determinants of Auditor Expertise , 1990 .

[58]  Jaime G. Carbonell,et al.  Machine Learning: A Historical and Methodological Analysis , 1983, AI Mag..

[59]  Timothy B. Bell,et al.  Empirical Analysis of Audit Uncertainty Qualifications , 1991 .

[60]  Charalambos Spathis,et al.  Auditee and audit firm characteristics as determinants of audit qualifications: Evidence from the Athens stock exchange , 2006 .

[61]  E. Altman.  Corporate financial distress : a complete guide to predicting, avoiding, and dealing with bankruptcy , 1983 .