Implications of Prior Probability Elicitation on Auditor Sample Size Decisions
研究不同先验概率引出方法如何影响审计师选择样本量的决策，帮助判断贝叶斯模型在实际审计中是否可行。
During the past decade, a number of researchers have applied a Bayesian or decision-theoretic approach to the study of auditing problems. Such research is largely normative in nature, suggesting how the auditor should behave. Little work has been done, however, to determine under what conditions such models can be operationalized. In order to employ a Bayesian model in practice, the auditor(s) must be able to express probabilities and to specify appropriate loss functions. The impact of alternative loss functions on audit decisions is reported by Lewis [1980]. The purpose of the study reported on here was to examine the effects of using different elicitation techniques on a subsequent auditing decision, that of sample size selection.