Hypothesis Testing
Null hypothesis: what we assume is true. Alternative hypothesis: what you are trying to show.
Developing a test
- Choose an estimator for the statistic you are working with
- Give the form of the test
Ex: reject the null hypothesis if our sample mean is greater than some value c.
- Find c
- Give a conclusion
Errors
- Type 1
The null hypothesis is true but you reject it.
Probability of making this error is Denoted alpha, and called level of significance or size of the test.
- Type 2
The null hypothesis is false but you fail to reject it.