a Bayesian interpretation of FDRs?

This week, I happened to re-read John Storey’ 2003 “The positive discovery rate: a Bayesian interpretation and the q-value”, because I wanted to check a connection with our testing by mixture [still in limbo] paper. I however failed to find what I was looking for because I could not find any Bayesian flavour in the paper apart from an FRD expressed as a “posterior probability” of the null, in the sense that the setting was one of opposing two simple hypotheses. When there is an unknown parameter common to the multiple hypotheses being tested, a prior distribution on the parameter makes these multiple hypotheses connected. What makes the connection puzzling is the assumption that the observed statistics defining the significance region are independent (Theorem 1). And it seems to depend on the choice of the significance region, which should be induced by the Bayesian modelling, not the opposite. (This alternative explanation does not help either, maybe because it is on baseball… Or maybe because the sentence “If a player’s [posterior mean] is above .3, it’s more likely than not that their true average is as well” does not seem to appear naturally from a Bayesian formulation.) [Disclaimer: I am not hinting at anything wrong or objectionable in Storey’s paper, just being puzzled by the Bayesian tag!]

Takaisin helsinkiin

I am off tomorrow morning to Helsinki for the European Meeting of Statisticians (EMS 2017). Where I will talk on how to handle multiple estimators in Monte Carlo settings (although I have not made enough progress in this direction to include anything truly novel in the talk!) Here are the slides:

I look forward this meeting, as I remember quite fondly the previous one I attended in Budapest. Which was of the highest quality in terms of talks and interactions. (I also remember working hard with Randal Douc on a yet-unfinished project!)

Mom on a tablet!

This X’mas, we decided to connect my mother to the Internet, as she was getting increasingly bothered by various local and national administrations for not filling forms on-line and for preferring agent-based to web-based support… She was also complaining of not seeing our family pictures any longer, except when we were sending her a USB key for her digital photo frame. So we got an Internet connection package from the local provider, along with a Samsung tablet that was offered as part of the package. Terrific! The connection started working almost instantly, using the tablet proved very manageable for my mother who had never touched a computer before (and who cannot handle the DVD player on her own!), and she sent the first email of her life! As well as looked on google map for a restaurant on a local beach town for the first time! (Sadly most restaurants were closed for the winter… We still managed to have a superb shellfish lunch in Ouistreham, near the ferry harbour.) I think this was a better solution than providing my mother with a computer, since she will only use the browsing, Skyping, and emailing facilities of the machine, relying on USB keys to download photographs. A few hours after starting the whole process, I am still amazed at how easily she got into the thing.