An Expert System Approach to Audit Planning and Evaluation in the Belief-Function Framework

Abstract

The main purpose of this article is to discuss an expert system approach for audit planning and evaluation using belief functions. First, we show how to use belief functions to represent strengths of various types of evidence such as positive, negative, or mixed items of evidence in an audit. The situation where one item of evidence relates to more than one audit objective or account is of special interest here, particularly the evidence that provides different levels of support to different audit objectives or accounts. Second, we illustrate the use of Auditor's Assistant, an expert system shell, for planning and evaluation of an audit. For this illustration, we use an actual audit program of one of the Big Six accounting firms for the accounts receivable account of a health care unit. Third, two types of sensitivity analyses are performed on the evidential network (audit program) created above. The first one determines the effects of the location of evidence in the network and the second one deals with the effects of variations in the auditor's judgment of the strength of evidence on the overall belief on each variable in the network. Finally, practical implications of the findings of the sensitivity analyses are discussed in the article.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 39 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.