Control the type 1 error

#1
Hello, everyone!

In my reasearch I have two primary endpoints - overall survival and progression-free survival. I test the superiority hypothesis H0: hazard ratio more than superiority margin and Ha is hazard ratio is less than superiority margin. And my question is: What method of control type I error may used if I want to prove two hypotheses of superiority at the same time (simultaneously)?

Many thanks!
 

hlsmith

Less is more. Stay pure. Stay poor.
#2
If you want to be cautious you just increase the alpha level, so a wider confidence interval needs to exclude the margin.