As much as I love Scotland, or because of it, I would not dream of suggesting to Scots that one side of the referendum sounds better than the other. However, I am rather annoyed at the yoyo-like reactions to the successive polls about the result, because, just like during the US elections, each poll is analysed separately rather than being pooled with the earlier ones in a reasonable meta-analysis… Where is Nate Silver when we need him?!
Archive for Glasgow
Deborah Mayo wrote a Saturday night special column on our Big Bayes stories issue in Statistical Science. She (predictably?) focussed on the critical discussions, esp. David Hand’s most forceful arguments where he essentially considers that, due to our (special issue editors’) selection of successful stories, we biased the debate by providing a “one-sided” story. And that we or the editor of Statistical Science should also have included frequentist stories. To which Deborah points out that demonstrating that “only” a frequentist solution is available may be beyond the possible. And still, I could think of partial information and partial inference problems like the “paradox” raised by Jamie Robbins and Larry Wasserman in the past years. (Not the normalising constant paradox but the one about censoring.) Anyway, the goal of this special issue was to provide a range of realistic illustrations where Bayesian analysis was a most reasonable approach, not to raise the Bayesian flag against other perspectives: in an ideal world it would have been more interesting to get discussants produce alternative analyses bypassing the Bayesian modelling but obviously discussants only have a limited amount of time to dedicate to their discussion(s) and the problems were complex enough to deter any attempt in this direction.
As an aside and in explanation of the cryptic title of this post, Deborah wonders at my use of endemic in the preface and at the possible mis-translation from the French. I did mean endemic (and endémique) in a half-joking reference to a disease one cannot completely get rid of. At least in French, the term extends beyond diseases, but presumably pervasive would have been less confusing… Or ubiquitous (as in Ubiquitous Chip for those with Glaswegian ties!). She also expresses “surprise at the choice of name for the special issue. Incidentally, the “big” refers to the bigness of the problem, not big data. Not sure about “stories”.” Maybe another occurrence of lost in translation… I had indeed no intent of connection with the “big” of “Big Data”, but wanted to convey the notion of a big as in major problem. And of a story explaining why the problem was considered and how the authors reached a satisfactory analysis. The story of the Air France Rio-Paris crash resolution is representative of that intent. (Hence the explanation for the above picture.)
In a “crazy travelling week” (dixit my daughter), I gave a talk at an IYS 2013 conference organised by Stephen Senn (formerly at Glasgow) and colleagues in the city of Luxembourg, Grand Duché du Luxembourg. I enjoyed very much the morning train trip there as it was a misty morning, with the sun rising over the frosted-white countryside. (I cannot say much about the city of Luxembourg itself though as I only walked the kilometre from the station to the conference hotel and the same way back. There was a huge gap on the plateau due to a river in the middle, which would have been a nice place to run, I presume…)
One of the few talks I attended there was about an econometric model with instrumental variables. In general, and this dates back to my student’s years at ENSAE, I do not get the motivation for the distinction between endogenous and exogenous in econometrics models. Especially in non-parametric models as, if we do not want to make parametric assumptions, we have difficulties in making instead correlation hypotheses… My bent would be to parametrise everything under the suspicion of this everything being correlated with everything. The instrumental variables econometricians seem so fond of appear to me like magical beings, since we have to know they are instrumental. And because they seem to allow to always come back to a linear setting, by eliminating the non-linear parts. Sounds like a “more for less” free-lunch deal. (Any pointer would be appreciated.) The speaker there actually acknowledged (verbatim) that they are indeed magical and that they cannot be justified by mathematics or statistics. A voodoo part of econometrics then?!
A second talk that left me perplexed was about a generalised finite mixture model. The model sounded like a mixture along time of individuals, ie a sort of clustering of longitudinal data. It looked like it should be easier to estimate than usual mixtures of regressions because an individual contributed to the same regression line for all the times when it was observed. The talk was uninspiring as it missed connections to EM and to Bayesian solutions, focussing instead on a gradient method that sounded inappropriate for a multimodal likelihood. (Funny enough, the choice in the number of regressions was done by BIC.)
Another book I received as a bedside gift at The Hospital, this one a gift from Magali. Waterline by Ross Raisin is the story of a Glaswegian former shipyard worker falling into a mental abyss of denial and grief after her wife died. She died from asbestos-related cancer, whose fibres were brought from the shipyard in Mick’s clothes. Mick seeks solitude and shuns contact with former colleagues and friends, seeing their return to a normal life as an aggression against his wife’s memory. When Mick cannot stay longer in his rented council house in Glasgow, fearing eviction and with no money left, he moves to London where he first finds a job as a dishwasher that alienates him even further into a bubble where he can cut others out. He is fired after a while for taking a very passive part in a local union and within a few weeks he spirals down into homelessness. The second part of the book sees him getting out very slowly and very reluctantly out of this state, with no clear sign of any return to (whatever we could call) normality…
This is a far from perfect book and the second part feels contrived, with a sort of “happy ending” out of the bottom end of Mick’s life. Still, Waterline is a strong book that marked me because it left me with a strong impression that the same fall could happen to any of us, under the right (or rather wrong) … Raisin has a highly convincing way of describing the inner mental paths taken by Mick to stop seeing others, including his children, for not returning to his uncertain cab driver job, and for giving up too readily looking for jobs, shelters, or help. The homeless-ness pages are terrifying in their somehow warped version of normality, when the “lives” of Mick and of his (leading) companion of misfortune Brian follow some kind of predestined pattern, from avoiding security guards by moving around to finding accessible toilets, to stealing food from delivery trucks, to tracking protected places to sleep away from the rough weather. At times, this feels like too much and too long, but I would support the idea that this is “exactly” (given my total lack of expertise in the matter…) reflecting the experience of those homeless men, with no future further than the next night or the next meal. The second part of the book also shows how hard it is to reconcile with a “normal” life, incl. a poignant chapter on Mike reuniting with his son uncomprehending why his father had not called for help. A highly recommended read, if not exactly on the brightest side of life…
As I was crossing the street, on my way to Institut Henri Poincaré to attend the Big’MC seminar with talks by Yves Atchadé on confidence intervals on MCMC ouput and Omiros Papaspiliopoulos on exact filtering, I thought the Panthéon had a nice enough background to deserve a picture. I also stopped by a nearby art shop to buy 0.7mm leads for my mechanical pencil and ended up discussing Charles Rennie Mackintosh with the seller, as I was wearing my University of Glasgow sweatshirt…
We have now completed our revision of the paper Relevant statistics for Bayesian model choice, written with Judith Rousseau, Jean-Michel Marin, and Natesh Pillai. It has been resubmitted to Series B and reposted on arXiv. The major change in the paper is the inclusion of a check about the relevance of a given summary statistics, as already explained in the talks I presented in Bristol and Glasgow. We also ran a realistic (and, I think, illuminating!) experiment to assess the impact of using one or two (δμ)² statistics as summaries in a simple population experiment, along with a theoretical explanation of the difference between both cases. This methodological addition answers in my opinion the major criticism contained in the review and I thus hope we can envision the eventual publication of this paper… In any case, the reviews have been tremendously helpful in improving the paper.