News Focus
News Focus
icon url

umiak

12/29/12 8:46 AM

#6418 RE: Rawnoc #6417

Yeah statistics can be strange, at least from some of the incomprehensible
posts I have seen regarding heat study. Not having the training, I tend
to rely on company filings/public comments, common sense and my bs detector.

Thanks for your effort, it certainly supports approval and points to
a possible, blowout. Very eager for data.
icon url

errett

12/29/12 11:05 AM

#6419 RE: Rawnoc #6417

Rawnoc---recall this from Siavoche's blog:

"As some of you know, many (including Mangrove Partners' Nathaniel August and several others) have tried "modeling" the HEAT study using enrollment data and events recorded at the interim analysis date and now for 380. According to some, if the control arm is truly 12 months, the separation in the curves could be fairly large due to unconfirmed guesses/estimates by many that the pooled median for the entire trial is in the 20-25 month range. I try not to speculate at all in this area (everything is purely just that, speculation based on fancy excel spreadsheets), as it is not an area of expertise of mine nor do I think such models carry much utility given all the variables that need to be accounted for. Sorry for the digression on this point."