I am performing a series of sequential hypothesis tests. I'm using the Bonferroni inequality, and a global alpha, to calculate alpha for each of the tests. This will assure me that my Type I error will be reasonable. However, as my alpha gets smaller, by beta will grow larger. And I do not want that to happen, since I may end up never rejecting the null hypothesis. In which ways can I control the Type II error (as well as Type I error) in a case like this?
Thanks a lot!
Thanks a lot!