A Continuous Constrained Optimization Model for Audit Sampling.
提出一种在合规性测试中选择最优审计抽样计划的方法，将可接受的β风险作为约束条件，最小化第一类错误和抽样的总成本，并考察了结果对输入误设的敏感性。
Abstract This article describes a method of selecting optimal audit sampling plans for compliance testing. The method requires the auditor to select an acceptable β risk, the risk of improper acceptance or of overreliance on internal control. This risk is then treated as a constraint in choosing a sampling plan that minimizes total cost of type 1 errors and sampling. The applicability of this approach to compliance testing is demonstrated. Previous approaches to selecting a sampling plan are shown to be sensitive to misspecification of the acceptable error rate. A method is developed that allows the auditor to consider costs and risks over the full range of acceptable values for this error rate. Methods for finding the optimal sampling plan are derived, and the sensitivity of results to input misspecifications is investigated.