Analytical procedures using artificial intelligence methods and rotated residual plot

Abstract

Auditors commonly use analytical procedures to analyze and evaluate financial information by examining relationships among financial and nonfinancial data for plausibility. These analytical procedures play an important role in the preliminary audit process. In addition, our auditing standards require analytical procedures as one of general auditing procedures from 1991. Much audit research reports a large number of errors encountered during an audit engagement were detected using analytical procedures. Despite the importance of analytical procedures, however, the knowledge and practicable methods about analytical procedures are quite sparse. In this study, we attemp to improve analytical methods by applying Artificial Intelligence (AI) methods including artificial neural networks (ANN) and case-based reasoning (CBR), and to perform pattern recognition of the investigation signals generated by analytical procedures. Five years of audited financial data from a large-sized firm were used to calculate four commonly applied financial ratios. Rotated residual plot (RRP) methods, not yet used in auditing, were introduced and utilized to improve the effectiveness of analytical procedures, first. This exploratory study shows that the use of AI methods to analyze patterns of related fluctuations across numerous financial ratios provides a more reliable indication to detect errors. In addition, AI methods using RRP values produce noticeably more effectiveness in detecting errors than traditional analytical procedures. From the results of this study, we suggest that the use of AI methods utilizing rotated residual plot (RRP) as a supplement to traditional analytical procedures will offer improved performance in recognizing material misstatements within the financial accounts.