But the big power calc question is not if DNDN used to large a HR, but clark's question about applying a constant risk model at all here (in the power calc analysis).
Many of the now censored 50? events will occur in the "meat" of the curve where you both have good seperation and sufficient events. None would have occured in the early months nbefore seperation.
I would not be surprised if one had a "real" curve and did a sim on it, they lost more power because of this than the 2%.
Regardless,
1) I do hope it (the final) hits a solid p value. 2) Management is not very clear what is going on and why.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.