News Focus
News Focus
Followers 843
Posts 122806
Boards Moderated 10
Alias Born 09/05/2002

Re: dewophile post# 343

Sunday, 08/06/2006 5:27:09 AM

Sunday, August 06, 2006 5:27:09 AM

Post# of 19309
A minor semantic matter re the KyberSept data:

>the statistically significant finding in the no-heparin group<

Because there were two data looks (28 days and 90 days) and two subgroups (heparin and no-heparin), I refrain from calling the 90-day results in the KyberSept no-heparin subgroup “statistically significant” even though the p-value was less than 0.05.

Given an aggregate alpha of 0.05 and two data looks on each of two subgroups, the p-value threshold for statistical significance of a single subgroup at a single data look is 0.0125, and that threshold was not met.

The above is academic, of course. The real issue is whether the data warrant further clinical development, and GTC and Leo have decided that they do.

“The efficient-market hypothesis may be
the foremost piece of B.S. ever promulgated
in any area of human knowledge!”

Discover What Traders Are Watching

Explore small cap ideas before they hit the headlines.

Join Today