Statistic Study Notes

1. hypothesis test: for data that meets certain prerequisites, make a hypothesis H0 first, and make a alternative hypothesis H1 (usually the opposite side of H0, or the case where H0 is not included ), A certain formula is used to calculate a value (for example, the square test is the square value). The theoretical value of this value should be 0;

The closer the theoretical value is, the more satisfied the hypothesis to accept the original hypothesis. The more deviated from the theoretical value, the less satisfied the original hypothesis. A threshold value can be set here;

P-value is calculated based on the formula used to calculate the test value. It is generally complicated and can be directly obtained through the Table query. (generally, p-value can be viewed as the reciprocal of the test value. If the test value is larger, the P-value is smaller. If the test value is smaller, the P-value is larger)

The actual hypothesis test should be clear, with the goal of proving whether to accept the original hypothesis or alternative hypothesis. The former requires a small test value and a large P value; the latter requires a large test value and a small P value.