There’s more intelligence in that one post by BI than in 1,000 posts by the DNDN bulls, IMO.
I argued similarly to BI in a thread a couple of months ago that was started by Ocyan’s contention on InvestorVillage that the FDA can be expected to give a sponsor a “free” implicit 0.05 of p-value for survival even if zero p-value is formally allocated to survival in the trial’s statistical-analysis plan. Clearly, if this were true—i.e., if the FDA routinely gave sponsors 0.05 of free p-value for survival—it would be wasteful for companies to allocate some of the scarce explicit p-value to survival and no clinical trial would do it. The fact that many clinical trials do allocate explicit p-value to survival is evidence that Ocyan’s contention is bogus.
BI is also correct in stating that 9902a was not supportive of 9901, IMO. No one—not even DNDN itself—stated a priori that “supportive” meant a low p-value using Cox regression analysis.
“The efficient-market hypothesis may be
the foremost piece of B.S. ever promulgated
in any area of human knowledge!”