An Analysis of Alternative Error Assumptions in Modeling the Reliability of Accounting Systems
检验了会计系统可靠性模型中关于误差相互依赖的不同简化假设对估计结果的影响，为审计师评估内部控制可靠性提供参考。
The reliability of internal control systems is a primary concern of auditors (AU, sec. 320). As a basis for improving assessments of reliability, accounting researchers have experimented with the development and use of formal models of the reliability of an accounting system (see Yu and Neter [1973] and Cushing [1974] for early examples). A primary objective of these models is to provide a less-subjective technique for assessing system reliability and the degree of audit reliance to be placed on internal controls. One common criticism of these models is that they tend to be based on restrictive assumptions about actual accounting environments.' Of course, models based on restrictive assumptions may still yield robust results, at least from an empirical viewpoint. This paper addresses this issue directly by examining the effect on the estimates derived from reliability models of various simplifying assumptions made about the interdependence of accounting errors. The next section of this paper discusses the error assumptions which are implicit in all reliability models. In section 3, a reliability model is developed which incorporates two alternative error assumptions. Then, in section 4, the prediction results of the reliability model under each