The New Statistics?


A group has “claimed to find a better way”:http://www.slaterfund.com/slaterfund/content_template.asp?file=newsdetail.asp&newsID=17 of doing statistics in clinical trials. It is based on a “pure likelihood”:http://64.233.161.104/search?q=cache:gmTfSWoOwSQJ:igitur-archive.library.uu.nl/dissertations/2004-0301-095707/c7.pdf+%22clinical+trials%22+%22pure+likelihood%22&hl=en&gl=us&ct=clnk&cd=4 approach, which, at first pass, seems to take the size of the p-value into account as well as whether it is below 0.05. I’ll have to investigate this a little more closely, but I’m a little skeptical at this point. I’ll have to see how they address such thorny issues as multiple comparisons.

Aside from that, we’ll have to see whether someone can run a business model based on this version of statistics. Bayesian statistics has been around longer, and I don’t see too many consultants, save the “Berrys”:http://berryconsultants.com/, able to base a business model on it. (I haven’t seen any more Bayesian clinical trial consultants with huge standing, at least.) This will change, of course, and the new methodology might catch on as well. I think it’ll take a long time, though.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: