小子样精度鉴定技术研究

Research on Accuracy Evaluation in Small Sample Case

讨论了小子样下的Bayes精度鉴定技术研究的一般方法，分析了考虑验前信息可信度时方差的检验，检验分为简单假设和复杂假设2种情况。给出了临界区域以及犯弃真和采伪两类错误的概率。仿真结果表明了方法的合理性。

This paper discusses the general method of Bayesian accuracy evaluation in small sample case, analyzes Analysis is presented on variance test when taking into consideration the prior credibility in the instances of complicated hypothesis and simple hypothesis. The critical region and the probability of Type Ⅰ and Type Ⅱerrors are given. Simulation results show that the method proposed in this paper is effective.