论文部分内容阅读
在数理统计的教学和研讨中,通常为了对总体的某些统计性质作出说明,或对几个总体的统计性质进行比较,即这些统计性质未知或不完全知道。此时可对总体作出一种“假设”。通过对样本的检验,来判断“假设”是否合理,从而确定假设是相容还是否定。并据此对总体统计性质作出说明。本文主要论述参数检验中易混淆的两个问题。 (一)关于两类错误当我们检验Ho时。总希望做到:如Ho为真就接受它,如Ho为假就拒绝它。但由于我们对全及总体期望μ并不掌握,因此有可能出现两类错误。 1、当Ho为真时,但一次抽样发生了小概率事件(其概率为a)因而否定Ho,这就犯了“以真为假”的错误,即
In the teaching and discussion of mathematical statistics, it is usually necessary to explain some of the statistical properties of the population as a whole or to compare the statistical properties of several populations as a whole, that is, the statistical properties are unknown or not fully known. At this point we can make a “hypothetical” for the population as a whole. Through the sample test, to determine “hypothesis ” is reasonable, to determine the assumptions are compatible or negative. Based on which the overall statistical nature is explained. This article mainly discusses two problems confusing in parametric test. (A) About two types of mistakes When we test Ho. Always want to do: If Ho is true to accept it, such as Ho to reject it. However, two types of mistakes are possible because we are not in a position to grasp the overall expectation. 1, when Ho is true, but a sampling of a small probability of occurrence of events (the probability of a) and thus denied Ho, which made the “false”, ie