Hypothesis testing learning notes

1. The basic concept of hypothesis testing

1.1. The null hypothesis （ Null hypothesis ）

The assumption made about the distribution of the population is Express , It is also called original hypothesis or null hypothesis

When the overall distribution type is known , Statistical assumptions involving only unknown parameters in the population distribution , It is called parameter hypothesis
When the overall distribution type is unknown , Statistical assumptions about the type of population distribution or some characteristics of population distribution , It is called nonparametric hypothesis

1.2. Two types of errors in hypothesis testing  

 

2. Significance test of single normal population parameter

  Yes, hypothesis A test of depends entirely on small probability events A The choice of .

2.1. μ test  

  When the variance is known, the significance test of the mean value of a normal population can be summarized into the following steps ：

 2.2. t test

2.3. test

3. Significance test of two normal population parameters

3.1. F test  

Significance test table of normal population parameters

4. Nonparametric hypothesis test

Goodness of fit test

Pearson statistics  

 

To sum up, use Goodness of fit test to test the hypothesis about the overall distribution , Steps are as follows ：