I (Daniel) will be giving a Stan overview talk on Thursday, August 20, 7 pm.
Bob gave a talk there 3.5 years ago. My talk will be light and include where we’ve been and where we’re going.
P.S. If you make it, find me. I have Stan stickers to give out.
P.P.S. Stan is on twitter.
Not really a comment—just an FYI.
This NYT story seems relevant to many of the discussions on this blog.
Please feel free to delete this post.
http://www.nytimes.com/2015/08/18/upshot/how-to-know-whether-to-believe-a-health-study.html?abt=0002&abg=1
Bob:
Just read that, I guess better than most but it is almost impossible to know whether to believe a health study if you don’t have access to study records and data or at least a knowledge of the group that were involved – its (currently) all done behind closed doors.
One of the comments suggested Cochrane and here is my favorite paper from (one of) them (confirming the above concern)
Key paper:
Tom Jefferson, et al (of The Cochrane Collaboration). Risk of bias in industry-funded oseltamivir trials: comparison of core reports versus full clinical study reports http://bmjopen.bmj.com/content/4/9/e005253.full
The conclusion:
“This approach is not possible when assessing trials reported in journal publications, in which articles necessarily reflect post hoc reporting with a far more sparse level of detail. We suggest that when bias is so limiting as to make meta-analysis results unreliable, either it should not be carried out or a prominent explanation of its clear limitations should be included alongside the meta-analysis.”
Naive question: Is Stan used actively in Machine Learning (motivated by the location of the talk)? Are there any examples or case studies? I’d love to read them.