假设检验的三个决策方法

Three Decision-Making Methods in Hypothesis Testing

摘要: 文章介绍了假设检验决策的三种方法：临界值法、p值法和区间估计法。临界值法适合检验统计量手工计算时使用，临界值法通过显著性水平$\alpha$ 确定临界值与检验统计量的值比较大小做出决策；若检验统计量值的计算依赖于软件，适合使用p值做决策，p值小于显著性水平$\alpha$ ，拒绝原假设，否则接受原假设；区间估计法帮助深层次理解区间估计与假设检验间的关系，对于同一总体参数，其区间估计和假设检验都依赖于同一样本统计量的抽样分布，参数的置信区间为参数假设检验的接受域。

Abstract: The article introduces three methods for making decisions in hypothesis testing: critical value method, p-value method and interval estimation method. The critical value method is suitable for use when calculating statistical quantities manually. The critical value method determines the critical value based on the significance level α, and then compares the value of the test statistic with the critical value to make a decision. If the calculation of the test statistic value relies on software, it is appropriate to make decisions based on the p-value. If the p-value is less than the significance level α, the null hypothesis is rejected; otherwise, the null hypothesis is accepted. The interval estimation method helps to deeply understand the relationship between interval estimation and hypothesis testing, For the same overall parameter, both interval estimation and hypothesis testing rely on the sampling distribution of the same sample statistics. The confidence interval of the parameter serves as the acceptance domain for the parameter hypothesis test.