Hypothesis contrast methods are models used in statistical inference whose objective is to verify whether an estimate adapts to population values. In less abstract words, the objective of hypothesis contrast methods is to check if an estimate adapts to reality ‘reliably’.
The assumptions are called parametric hypotheses. That is, a decision criterion is established. If with this condition the reference hypothesis is accepted, then we can state with some probability that the estimate can be very close to the assumed real value.
In every hypothesis contrast there are two assumptions. The null hypothesis (H0) that reflects the idea that a value has a predetermined value. If the null hypothesis (H0) is rejected, then the alternative hypothesis (H1) is accepted .
Example of hypothesis contrast
Continuing with the previous results.
- Match A:32%
- Match B:51%
- Match C:17%
Contrast of hypotheses that A has a 32% to 95% confidence.
- H0:Does not have a 32% vote with a 95% probability
- H1:Has a 32% vote with a 95% chance
In the case that we can affirm that it will have 32%, in statistical terms, it is said that the null hypothesis (H0) is rejected. That is, it is rejected that no. Otherwise, it is accepted. That is, it is accepted that no.