The effectiveness of artificial neural networks applied to analytical procedures using high level data: a simulation analysis

Stewart Li, Richard Fisher and Michael Falta

Meditari Accountancy Research, 2020, vol. 29, issue 6, 1425-1450

Abstract: Purpose - Auditors are required to perform analytical procedures during the planning and concluding phases of the audit. Such procedures typically use data aggregated at a high level. The authors investigate whether artificial neural networks, a more sophisticated technique for analytical review than typically used by auditors, may be effective when using high level data. Design/methodology/approach - Data from companies operating in the dairy industry were used to train an artificial neural network. Data with and without material seeded errors were used to test alternative techniques. Findings - Results suggest that the artificial neural network approach was not significantly more effective (taking into account both Type I and II errors) than traditional ratio and regression analysis, and none of the three approaches provided more overall effectiveness than a purely random procedure. However, the artificial neural network approach did yield considerably fewer Type II errors than the other methods, which suggests artificial neural networks could be a candidate to improve the performance of analytical procedures in circumstances where Type II error rates are the primary concern of the auditor. Originality/value - The authors extend the work of Coakley and Brown (1983) by investigating the application of artificial neural networks as an analytical procedure using aggregated data. Furthermore, the authors examine multiple companies from one industry and supplement financial information with both exogenous industry and macro-economic data.

Keywords: Artificial neural networks; Effectiveness; Regression analysis; Analytical procedures; Ratios (search for similar items in EconPapers)
Date: 2020
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.emerald.com/insight/content/doi/10.110 ... d&utm_campaign=repec (text/html)
https://www.emerald.com/insight/content/doi/10.110 ... d&utm_campaign=repec (application/pdf)
Access to full text is restricted to subscribers

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eme:medarp:medar-06-2020-0920

DOI: 10.1108/MEDAR-06-2020-0920

Access Statistics for this article

Meditari Accountancy Research is currently edited by Prof Charl de Villiers and Warren Maroun

More articles in Meditari Accountancy Research from Emerald Group Publishing Limited
Bibliographic data for series maintained by Emerald Support ().