Baidu Encyclopedia version

Hypothesis testing is a method used to test statistical hypotheses in inferential statistics. The "statistical hypothesis" is a scientific hypothesis that can be tested by observing a model of a set of random variables. Once the unknown parameters can be estimated, it is desirable to make an appropriate inference of the unknown true parameter values ​​based on the results.

The statistical assumption of parameters is a discussion of one or more parameters. Where the null hypothesis is to be tested for correctness, the null hypothesis is usually determined by the investigator, reflecting the investigator's perception of the unknown parameters. The discussion of other relevant parameters relative to the null hypothesis is an alternative hypothesis, which usually reflects another (opposing) view of the possible value of the parameter by the investigator performing the test (in other words, the alternative hypothesis is usually It is the researcher most want to know).

see details

 

Wikipedia version

Statistical assumptions, sometimes referred to as verification data analysis, are a hypothesis that, based on a testable observation, a process model passes a set of random variables. Statistical hypothesis testing is a method of statistical inference.

Usually, two statistical data sets are compared, or a data set obtained by sampling is compared with a synthetic data set from an idealized model. A hypothesis is proposed for the statistical relationship between the two data sets, and it is used as an alternative to make a more idealized null hypothesis, and there is no relationship between the two data sets. If according to the threshold probability-significance level, the relationship between the data sets will be impossible to achieve the null hypothesis, then the comparison is considered statistically significant. Hypothesis testing is used to determine which results of the study will lead to rejection of the null hypothesis for a pre-specified significance level. The process of distinguishing between null hypotheses and alternative hypotheses is assisted by identifying errors of two conceptual types.

The first type occurs when the null hypothesis is incorrectly rejected. When the null hypothesis is incorrectly assumed to be true, the second type of error (type 1 and type 2 errors) occurs. By specifying a threshold probability ('alpha'), for example, the allowable risk of producing a type 1 error, the statistical decision-making process can be controlled.

see details

Easyai public number