Mon 22 Aug, 9:50am, at Harvard Science Center Hall A:

Taking Bayesian Inference Seriously

Over the years I have been moving toward the use of informative priors in more and more of my applications. I will discuss several examples from theory, application, and computing where traditional noninformative priors lead to disaster, but a little bit of prior information can make everything work out. Informative priors also can resolve some of the questions of replication and multiple comparisons that have recently shook the world of science. It’s funny for me to say this, after having practiced Bayesian statistics for nearly thirty years, but I’m only now realizing the true value of the prior distribution.

Have recently shaken.

Any chance you can post the talk.

Thanks,

-Roy

Roy,

OK, sure, I’ll link, but the slides aren’t so special, they’re pretty much a mash-up of material from various talks from the past few years. The real innovation is in the delivery, I think.

Maybe it will show up here..?

http://cmsa.fas.harvard.edu/category/videos/

I’d be interested to see your paper. (I have a paper called “Taking Bayesian Criticisms Seriously”–not relevant, just thought I’d mention it.)

Do you have a link to your paper?

What’s the connection to “big data”? (I think I can guess, but I thought I’d go with a wh-question this time rather than speculate.)

An alternative to the Bayesian testing is the chi-square goodness of fit test for proportions when the population priors are well known. I would appreciate a discussion on the differences between these two approaches.