假设检验与参数英文概念解释汇集.docVIP

Hypothesis testing Average could not satisfy our data analysis. Because there may exist error, and many reason would result the error. So we need hypothesis testing to judge whether the error would affect the result. Hypothesis test is the method that makes a hypothesis based on the probability distribution or distribution parameter of population, according to the sample observations, makes use of statistical methods of analysis to test the correctness of this hypothesis, and decides to accept or reject the hypothesis. Hypothesis test also is called the test of significance. Hypothesis testing is another important content in statistics. There are many methods in hypothesis testing, such as t-inspection, F-inspection,-inspection and so on. Hypothesis testing can judge the different between estimated value and general value with a little risk. It estimated the total by the sample, and the result is reliable entirely. Besides, the purpose of inspection is to analyse if evident contrast between t sample index and total index, instead of doubt computed result of sample index itself. From this meaning, hypothesis testing can also named significance test. Hypothesis testing is a statistical inference method, which is used to determine whether the difference between sample and sample or sample and population caused by sampling error or essential difference. The basic principle of hypothesis testing is to make certain hypothesis for the character of population, and then reject or accept this hypothesis through inferences reasoning. And there are three basic steps in hypothesis testing that are: ① set up hypothesis and determine the size of a test, while, the hypothesis contains null hypothesis （H0）and alternative hypothesis (HA). The meaning of null hypothesis is that the test surface difference caused by the sampling error, while, the alternative hypothesis shows that the test surface difference also contains test real difference besides sampling error. ② select test meth