Archive for Bayesian model choice

JSM 2014, Boston [#3]

Posted in Statistics, University life with tags , , , , , , , on August 8, 2014 by xi'an

Today I gave a talk in the Advances in model selection session. Organised by Veronika Rockova and Ed George. (A bit of pre-talk stress: I actually attempted to change my slides at 5am and only managed to erase the current version! I thus left early enough to stop by the presentation room…) Here are the final slides, which have much in common with earlier versions, but also borrowed from Jean-Michel Marin’s talk in Cambridge. A posteriori, I think the talk missed one slide on the practical run of the ABC random forest algorithm, since later questions showed miscomprehension from the audience.

The other talks in this session were by Andreas Buja [whom I last met in Budapest last year] on valid post-modelling inference. A very relevant reflection on the fundamental bias in statistical modelling. Then by Nick Polson, about efficient ways to compute MAP for objective functions that are irregular.  Great entry into optimisation methods I had never heard of earlier.! (The abstract is unrelated.) And last but not least by Veronika Rockova, on mixing Indian buffet processes with spike-and-slab priors for factor analysis with unknown numbers of factors. A definitely advanced contribution to factor analysis, with a very nice idea of introducing a non-identifiable rotation to align on orthogonal designs. (Here too the abstract is unrelated, a side effect of the ASA requiring abstracts sent very long in advance.)

Although discussions lasted well into the following Bayesian Inference: Theory and Foundations session, I managed to listen to a few talks there. In particular, a talk by Keli Liu on constructing non-informative priors. A question of direct relevance. The notion of objectivity is to achieve a frequentist distribution of the Bayes factor associated with the point null that is constant. Or has a constant quantile at a given level. The second talk by Alexandra Bolotskikh related to older interests of mine’s, namely the construction of improved confidence regions in the spirit of Stein. (Not that surprising, given that a coauthor is Marty Wells, who worked with George and I on the topic.) A third talk by Abhishek Pal Majumder (jointly with Jan Hanning) dealt on a new type of fiducial distributions, with matching prior properties. This sentence popped a lot over the past days, but this is yet another area where I remain puzzled by the very notion. I mean the notion of fiducial distribution. Esp. in this case where the matching prior gets even closer to being plain Bayesian.

R/Rmetrics in Paris [alas!]

Posted in Mountains, pictures, R, Statistics, Travel, University life with tags , , , , , , , , , , , , , , , , , , , on June 30, 2014 by xi'an

Bernard1Today I gave a talk on Bayesian model choice in a fabulous 13th Century former monastery in the Latin Quarter of Paris… It is the Collège des Bernardins, close to Jussieu and Collège de France, unbelievably hidden to the point I was not aware of its existence despite having studied and worked in Jussieu since 1982… I mixed my earlier San Antonio survey on importance sampling approximations to Bayes factors with an entry to our most recent work on ABC with random forests. This was the first talk of the 8th R/Rmetrics workshop taking place in Paris this year. (Rmetrics is aiming at aggregating R packages with econometrics and finance applications.) And I had a full hour and a half to deliver my lecture to the workshop audience. Nice place, nice people, new faces and topics (and even andouille de Vire for lunch!): why should I complain with an alas in the title?!Bernard2What happened is that the R/Rmetrics meetings have been till this year organised in Meielisalp, Switzerland. Which stands on top of Thuner See and… just next to the most famous peaks of the Bernese Alps! And that I had been invited last year but could not make it… Meaning I lost a genuine opportunity to climb one of my five dream routes, the Mittelegi ridge of the Eiger. As the future R/Rmetrics meetings will not take place there.

A lunch discussion at the workshop led me to experiment the compiler library in R, library that I was unaware of. The impact on the running time is obvious: recycling the fowler function from the last Le Monde puzzle,

> bowler=cmpfun(fowler)
> N=20;n=10;system.time(fowler(pred=N))
   user  system elapsed 
 52.647   0.076  56.332 
> N=20;n=10;system.time(bowler(pred=N))
   user  system elapsed 
 51.631   0.004  51.768 
> N=20;n=15;system.time(bowler(pred=N))
   user  system elapsed 
 51.924   0.024  52.429 
> N=20;n=15;system.time(fowler(pred=N))
   user  system elapsed 
 52.919   0.200  61.960 

shows a ten- to twenty-fold gain in system time, if not in elapsed time (re-alas!).

Bayesian Data Analysis [BDA3 - part #2]

Posted in Books, Kids, R, Statistics, University life with tags , , , , , , , , on March 31, 2014 by xi'an

Here is the second part of my review of Gelman et al.’ Bayesian Data Analysis (third edition):

“When an iterative simulation algorithm is “tuned” (…) the iterations will not in general converge to the target distribution.” (p.297)

Part III covers advanced computation, obviously including MCMC but also model approximations like variational Bayes and expectation propagation (EP), with even a few words on ABC. The novelties in this part are centred at Stan, the language Andrew is developing around Hamiltonian Monte Carlo techniques, a sort of BUGS of the 10’s! (And of course Hamiltonian Monte Carlo techniques themselves. A few (nit)pickings: the book advises important resampling without replacement (p.266) which makes some sense when using a poor importance function but ruins the fundamentals of importance sampling. Plus, no trace of infinite variance importance sampling? of harmonic means and their dangers? In the Metropolis-Hastings algorithm, the proposal is called the jumping rule and denoted by Jt, which, besides giving the impression of a Jacobian, seems to allow for time-varying proposals and hence time-inhomogeneous Markov chains, which convergence properties are much hairier. (The warning comes much later, as exemplified in the above quote.) Moving from “burn-in” to “warm-up” to describe the beginning of an MCMC simulation. Being somewhat 90’s about convergence diagnoses (as shown by the references in Section 11.7), although the book also proposes new diagnoses and relies much more on effective sample sizes. Particle filters are evacuated in hardly half-a-page. Maybe because Stan does not handle particle filters. A lack of intuition about the Hamiltonian Monte Carlo algorithms, as the book plunges immediately into a two-page pseudo-code description. Still using physics vocabulary that put me (and maybe only me) off. Although I appreciated the advice to check analytical gradients against their numerical counterpart.

“In principle there is no limit to the number of levels of variation that can be handled in this way. Bayesian methods provide ready guidance in handling the estimation of the unknown parameters.” (p.381)

I also enjoyed reading the part about modes that stand at the boundary of the parameter space (Section 13.2), even though I do not think modes are great summaries in Bayesian frameworks and while I do not see how picking the prior to avoid modes at the boundary avoids the data impacting the prior, in fine. The variational Bayes section (13.7) is equally enjoyable, with a proper spelled-out illustration, introducing an unusual feature for Bayesian textbooks.  (Except that sampling without replacement is back!) Same comments for the Expectation Propagation (EP) section (13.8) that covers brand new notions. (Will they stand the test of time?!)

“Geometrically, if β-space is thought of as a room, the model implied by classical model selection claims that the true β has certain prior probabilities of being in the room, on the floor, on the walls, in the edge of the room, or in a corner.” (p.368)

Part IV is a series of five chapters about regression(s). This is somewhat of a classic, nonetheless  Chapter 14 surprised me with an elaborate election example that dabbles in advanced topics like causality and counterfactuals. I did not spot any reference to the g-prior or to its intuitive justifications and the chapter mentions the lasso as a regularisation technique, but without any proper definition of this “popular non-Bayesian form of regularisation” (p.368). In French: with not a single equation! Additional novelty may lie in the numerical prior information about the correlations. What is rather crucially (cruelly?) missing though is a clearer processing of variable selection in regression models. I know Andrew opposes any notion of a coefficient being exactly equal to zero, as ridiculed through the above quote, but the book does not reject model selection, so why not in this context?! Chapter 15 on hierarchical extensions stresses the link with exchangeability, once again. With another neat election example justifying the progressive complexification of the model and the cranks and toggles of model building. (I am not certain the reparameterisation advice on p.394 is easily ingested by a newcomer.) The chapters on robustness (Chap. 17) and missing data (Chap. 18) sound slightly less convincing to me, esp. the one about robustness as I never got how to make robustness agree with my Bayesian perspective. The book states “we do not have to abandon Bayesian principles to handle outliers” (p.436), but I would object that the Bayesian paradigm compels us to define an alternative model for those outliers and the way they are produced. One can always resort to a drudging exploration of which subsample of the dataset is at odds with the model but this may be unrealistic for large datasets and further tells us nothing about how to handle those datapoints. The missing data chapter is certainly relevant to such a comprehensive textbook and I liked the survey illustration where the missing data was in fact made of missing questions. However, I felt the multiple imputation part was not well-presented, fearing readers would not understand how to handle it…

“You can use MCMC, normal approximation, variational Bayes, expectation propagation, Stan, or any other method. But your fit must be Bayesian.” (p.517)

Part V concentrates the most advanced material, with Chapter 19 being mostly an illustration of a few complex models, slightly superfluous in my opinion, Chapter 20 a very short introduction to functional bases, including a basis selection section (20.2) that implements the “zero coefficient” variable selection principle refuted in the regression chapter(s), and does not go beyond splines (what about wavelets?), Chapter 21 a (quick) coverage of Gaussian processes with the motivating birth-date example (and two mixture datasets I used eons ago…), Chapter 22 a more (too much?) detailed study of finite mixture models, with no coverage of reversible-jump MCMC, and Chapter 23 an entry on Bayesian non-parametrics through Dirichlet processes.

“In practice, for well separated components, it is common to remain stuck in one labelling across all the samples that are collected. One could argue that the Gibbs sampler has failed in such a case.” (p.535)

To get back to mixtures, I liked the quote about the label switching issue above, as I was “one” who argued that the Gibbs sampler fails to converge! The corresponding section seems to favour providing a density estimate for mixture models, rather than component-wise evaluations, but it nonetheless mentions the relabelling by permutation approach (if missing our 2000 JASA paper). The section about inferring on the unknown number of components suggests conducting a regular Gibbs sampler on a model with an upper bound on the number of components and then checking for empty components, an idea I (briefly) considered in the mid-1990’s before the occurrence of RJMCMC. Of course, the prior on the components matters and the book suggests using a Dirichlet with fixed sum like 1 on the coefficients for all numbers of components.

“14. Objectivity and subjectivity: discuss the statement `People tend to believe results that support their preconceptions and disbelieve results that surprise them. Bayesian methods tend to encourage this undisciplined mode of thinking.’¨ (p.100)

Obviously, this being a third edition begets the question, what’s up, doc?!, i.e., what’s new [when compared with the second edition]? Quite a lot, even though I am not enough of a Gelmanian exegist to produce a comparision table. Well, for a starter, David Dunson and Aki Vethtari joined the authorship, mostly contributing to the advanced section on non-parametrics, Gaussian processes, EP algorithms. Then the Hamiltonian Monte Carlo methodology and Stan of course, which is now central to Andrew’s interests. The book does include a short Appendix on running computations in R and in Stan. Further novelties were mentioned above, like the vision of weakly informative priors taking over noninformative priors but I think this edition of Bayesian Data Analysis puts more stress on clever and critical model construction and on the fact that it can be done in a Bayesian manner. Hence the insistence on predictive and cross-validation tools. The book may be deemed somewhat short on exercices, providing between 3 and 20 mostly well-developed problems per chapter, often associated with datasets, rather than the less exciting counter-example above. Even though Andrew disagrees and his students at ENSAE this year certainly did not complain, I personally feel a total of 220 exercices is not enough for instructors and self-study readers. (At least, this reduces the number of email requests for solutions! Esp. when 50 of those are solved on the book website.) But this aspect is a minor quip: overall this is truly the reference book for a graduate course on Bayesian statistics and not only Bayesian data analysis.

Bayesian Data Analysis [BDA3]

Posted in Books, Kids, R, Statistics, University life with tags , , , , , , , , on March 28, 2014 by xi'an

Andrew Gelman and his coauthors, John Carlin, Hal Stern, David Dunson, Aki Vehtari, and Don Rubin, have now published the latest edition of their book Bayesian Data Analysis. David and Aki are newcomers to the authors’ list, with an extended section on non-linear and non-parametric models. I have been asked by Sam Behseta to write a review of this new edition for JASA (since Sam is now the JASA book review editor). After wondering about my ability to produce an objective review (on the one hand, this is The Competition  to Bayesian Essentials!, on the other hand Andrew is a good friend spending the year with me in Paris), I decided to jump for it and write a most subjective review, with the help of Clara Grazian who was Andrew’s teaching assistant this year in Paris and maybe some of my Master students who took Andrew’s course. The second edition was reviewed in the September 2004 issue of JASA and we now stand ten years later with an even more impressive textbook. Which truly what Bayesian data analysis should be.

This edition has five parts, Fundamentals of Bayesian Inference, Fundamentals of Bayesian Data Analysis, Advanced Computation, Regression Models, and Non-linear and Non-parametric Models, plus three appendices. For a total of xiv+662 pages. And a weight of 2.9 pounds (1395g on my kitchen scale!) that makes it hard to carry around in the metro…. I took it to Warwick (and then Nottingham and Oxford and back to Paris) instead.

We could avoid the mathematical effort of checking the integrability of the posterior density (…) The result would clearly show the posterior contour drifting off toward infinity.” (p.111)

While I cannot go into a detailed reading of those 662 pages (!), I want to highlight a few gems. (I already wrote a detailed and critical analysis of Chapter 6 on model checking in that post.) The very first chapter provides all the necessary items for understanding Bayesian Data Analysis without getting bogged in propaganda or pseudo-philosophy. Then the other chapters of the first part unroll in a smooth way, cruising on the B highway… With the unique feature of introducing weakly informative priors (Sections 2.9 and 5.7), like the half-Cauchy distribution on scale parameters. It may not be completely clear how weak a weakly informative prior, but this novel notion is worth including in a textbook. Maybe a mild reproach at this stage: Chapter 5 on hierarchical models is too verbose for my taste, as it essentially focus on the hierarchical linear model. Of course, this is an essential chapter as it links exchangeability, the “atom” of Bayesian reasoning used by de Finetti, with hierarchical models. Still. Another comment on that chapter: it broaches on the topic of improper posteriors by suggesting to run a Markov chain that can exhibit improperness by enjoying an improper behaviour. When it happens as in the quote above, fine!, but there is no guarantee this is always the case! For instance, improperness may be due to regions near zero rather than infinity. And a last barb: there is a dense table (Table 5.4, p.124) that seems to run contrariwise to Andrew’s avowed dislike of tables. I could also object at the idea of a “true prior distribution” (p.128), or comment on the trivia that hierarchical chapters seem to attract rats (as I also included a rat example in the hierarchical Bayes chapter of Bayesian Choice and so does the BUGS Book! Hence, a conclusion that Bayesian textbooks are better be avoided by muriphobiacs…)

“Bayes factors do not work well for models that are inherently continuous (…) Because we emphasize continuous families of models rather than discrete choices, Bayes factors are rarely relevant in our approach to Bayesian statistics.” (p.183 & p.193)

Part II is about “the creative choices that are required, first to set up a Bayesian model in a complex problem, then to perform the model checking and confidence building that is typically necessary to make posterior inferences scientifically defensible” (p.139). It is certainly one of the strengths of the book that it allows for a critical look at models and tools that are rarely discussed in more theoretical Bayesian books. As detailed in my  earlier post on Chapter 6, model checking is strongly advocated, via posterior predictive checks and… posterior predictive p-values, which are at best empirical indicators that something could be wrong, definitely not that everything’s allright! Chapter 7 is the model comparison equivalent of Chapter 6, starting with the predictive density (aka the evidence or the marginal likelihood), but completely bypassing the Bayes factor for information criteria like the Watanabe-Akaike or widely available information criterion (WAIC), and advocating cross-validation, which is empirically satisfying but formally hard to integrate within a full Bayesian perspective. Chapter 8 is about data collection, sample surveys, randomization and related topics, another entry that is missing from most Bayesian textbooks, maybe not that surprising given the research topics of some of the authors. And Chapter 9 is the symmetric in that it focus on the post-modelling step of decision making.

(Second part of the review to appear on Monday, leaving readers the weekend to recover!)

new MCMC algorithm for Bayesian variable selection

Posted in pictures, Statistics, Travel, University life with tags , , , , , , , , , on February 25, 2014 by xi'an

Flight from Bristol to Amsterdam, April 03, 2011Unfortunately, I will miss the incoming Bayes in Paris seminar next Thursday (27th February), as I will be flying to Montréal and then Québec at the time (despite having omitted to book a flight till now!). Indeed Amandine Shreck will give a talk at 2pm in room 18 of ENSAE, Malakoff, on A shrinkage-thresholding Metropolis adjusted Langevin algorithm for Bayesian variable selection, a work written jointly with Gersende Fort, Sylvain Le Corff, and Eric Moulines, and arXived at the end of 2013 (which may explain why I missed it!). Here is the abstract:

This paper introduces a new Markov Chain Monte Carlo method to perform Bayesian variable selection in high dimensional settings. The algorithm is a Hastings-Metropolis sampler with a proposal mechanism which combines (i) a Metropolis adjusted Langevin step to propose local moves associated with the differentiable part of the target density with (ii) a shrinkage-thresholding step based on the non-differentiable part of the target density which provides sparse solutions such that small components are shrunk toward zero. This allows to sample from distributions on spaces with different dimensions by actually setting some components to zero. The performances of this new procedure are illustrated with both simulated and real data sets. The geometric ergodicity of this new transdimensional Markov Chain Monte Carlo sampler is also established.

(I will definitely get a look at the paper over the coming days!)

MCMSki IV [day 3]

Posted in Mountains, pictures, R, Statistics, Travel, University life with tags , , , , , , , , , , , , , , on January 9, 2014 by xi'an

ridge5Already on the final day..! And still this frustration in being unable to attend three sessions at once… Andrew Gelman started the day with a non-computational talk that broached on themes that are familiar to readers of his blog, on the misuse of significance tests and on recommendations for better practice. I then picked the Scaling and optimisation of MCMC algorithms session organised by Gareth Roberts, with optimal scaling talks by Tony Lelièvre, Alex Théry and Chris Sherlock, while Jochen Voss spoke about the convergence rate of ABC, a paper I already discussed on the blog. A fairly exciting session showing that MCMC’ory (name of a workshop I ran in Paris in the late 90’s!) is still well and alive!

After the break (sadly without the ski race!), the software round-table session was something I was looking for. The four softwares covered by this round-table were BUGS, JAGS, STAN, and BiiPS, each presented according to the same pattern. I would have like to see a “battle of the bands”, illustrating pros & cons for each language on a couple of models & datasets. STAN got the officious prize for cool tee-shirts (we should have asked the STAN team for poster prize tee-shirts). And I had to skip the final session for a flu-related doctor appointment…

I called for a BayesComp meeting at 7:30, hoping for current and future members to show up and discuss the format of the future MCMski meetings, maybe even proposing new locations on other “sides of the Italian Alps”! But (workshop fatigue syndrome?!), no-one showed up. So anyone interested in discussing this issue is welcome to contact me or David van Dyk, the new BayesComp program chair.

MCMSki [day 2]

Posted in Mountains, pictures, Statistics, University life with tags , , , , , , , , , on January 8, 2014 by xi'an

ridge3I was still feeling poorly this morning with my brain in a kind of flu-induced haze so could not concentrate for a whole talk, which is a shame as I missed most of the contents of the astrostatistics session put together by David van Dyk… Especially the talk by Roberto Trotta I was definitely looking for. And the defence of nested sampling strategies for marginal likelihood approximations. Even though I spotted posterior distributions for WMAP and Plank data on the ΛCDM that reminded me of our own work in this area… Apologies thus to all speakers for dozing in and out, it was certainly not due to a lack of interest!

Sebastian Seehars mentioned emcee (for ensemble Monte Carlo), with a corresponding software nicknamed “the MCMC hammer”, and their own CosmoHammer software. I read the paper by Goodman and Ware (2010) this afternoon during the ski break (if not on a ski lift!). Actually, I do not understand why an MCMC should be affine invariant: a good adaptive MCMC sampler should anyway catch up the right scale of the target distribution. Other than that, the ensemble sampler reminds me very much of the pinball sampler we developed with Kerrie Mengersen (1995 Valencia meeting), where the target is the product of L targets,

\pi(x_1)\cdots\pi(x_L)

and a Gibbs-like sampler can be constructed, moving one component (with index k, say) of the L-sample at a time. (Just as in the pinball sampler.) Rather than avoiding all other components (as in the pinball sampler), Goodman and Ware draw a single other component at random  (with index j, say) and make a proposal away from it:

\eta=x_j(t) + \zeta \{x_k(t)-x_j(t)\}

where ζ is a scale random variable with (log-) symmetry around 1. The authors claim improvement over a single track Metropolis algorithm, but it of course depends on the type of Metropolis algorithms that is chosen… Overall, I think the criticism of the pinball sampler also applies here: using a product of targets can only slow down the convergence. Further, the affine structure of the target support is not a given. Highly constrained settings should not cope well with linear transforms and non-linear reparameterisations would be more efficient….

Follow

Get every new post delivered to your Inbox.

Join 640 other followers