multilevel linear models, Gibbs samplers, and multigrid decompositions

A paper by Giacommo Zanella (formerly Warwick) and Gareth Roberts (Warwick) is about to appear in Bayesian Analysis and (still) open for discussion. It examines in great details the convergence properties of several Gibbs versions of the same hierarchical posterior for an ANOVA type linear model. Although this may sound like an old-timer opinion, I find it good to have Gibbs sampling back on track! And to have further attention to diagnose convergence! Also, even after all these years (!), it is always a surprise  for me to (re-)realise that different versions of Gibbs samplings may hugely differ in convergence properties.

At first, intuitively, I thought the options (1,0) (c) and (0,1) (d) should be similarly performing. But one is “more” hierarchical than the other. While the results exhibiting a theoretical ordering of these choices are impressive, I would suggest pursuing an random exploration of the various parameterisations in order to handle cases where an analytical ordering proves impossible. It would most likely produce a superior performance, as hinted at by Figure 4. (This alternative happens to be briefly mentioned in the Conclusion section.) The notion of choosing the optimal parameterisation at each step is indeed somewhat unrealistic in that the optimality zones exhibited in Figure 4 are unknown in a more general model than the Gaussian ANOVA model. Especially with a high number of parameters, parameterisations, and recombinations in the model (Section 7).

An idle question is about the extension to a more general hierarchical model where recentring is not feasible because of the non-linear nature of the parameters. Even though Gaussianity may not be such a restriction in that other exponential (if artificial) families keeping the ANOVA structure should work as well.

Theorem 1 is quite impressive and wide ranging. It also reminded (old) me of the interleaving properties and data augmentation versions of the early-day Gibbs. More to the point and to the current era, it offers more possibilities for coupling, parallelism, and increasing convergence. And for fighting dimension curses.

“in this context, imposing identifiability always improves the convergence properties of the Gibbs Sampler”

Another idle thought of mine is to wonder whether or not there is a limited number of reparameterisations. I think that by creating unidentifiable decompositions of (some) parameters, eg, μ=μ¹+μ²+.., one can unrestrictedly multiply the number of parameterisations. Instead of imposing hard identifiability constraints as in Section 4.2, my intuition was that this de-identification would increase the mixing behaviour but this somewhat clashes with the above (rigorous) statement from the authors. So I am proven wrong there!

Unless I missed something, I also wonder at different possible implementations of HMC depending on different parameterisations and whether or not the impact of parameterisation has been studied for HMC. (Which may be linked with Remark 2?)

unimaginable scale culling

Despite the evidence brought by ABC on the inefficiency of culling in massive proportions the British Isles badger population against bovine tuberculosis, the [sorry excuse for a] United Kingdom government has permitted a massive expansion of badger culling, with up to 64,000 animals likely to be killed this autumn… Since the cows are the primary vectors of the disease, what about starting with these captive animals?!

MCqMC 2016 [#1]

This week, I attend the MCqMC 2016 conference in Stanford, which is quite an exciting gathering of researchers involved in various aspects of Monte Carlo methods. As Art Owen put it in his welcoming talk, the whole Carlo family is there! (Not to mention how pleasant the Stanford Campus currently is, after the scorching heat we met the past week in Northern California inlands.) My talk is on folded Markov chains, which is a proposal Randal and I have been working on for quite a while, with Gareth joining us more recently. The basic idea was inspired from a discussion I had about a blog post, so long ago that I cannot even trace it! Namely, when defining an inside set A and an outside set, such that the outside set can be projected onto the inside set, one can fold both the target and the proposal, essentially looking at a collection of values for each step of the Markov chain. In other words, the problem can be reduced to A at essentially no cost and with the benefits of a compact support A and of a possibly uniformly ergodic Markov chain. We are still working on the paper, but the idea is both cool and straightforward, so we decided to talk about it at Nordstat 2016 and now MCqMC 2016.

Series B news

The Journal of the Royal Statistical Society, Series B, has a new cover, a new colour and a new co-editor. As can be seen from the above shots, the colour is now a greenish ochre, with a picture of pedestrians on a brick plaza as a background, not much related to statistical methodology as far as I can tell. More importantly, the new co-editor for the coming four years is Piotr Fryzlewicz, professor at the London School of Economics, who will share the burden with Ingrid van Keilegom professor from UCL (Louvain-la-Neuve) who is now starting her third year… My friend, colleague and successor as Series B editor Gareth Roberts is now retiring after four years of hard work towards making Series B one of the top journals in Statistics. Thanks Gareth and best wishes to Ingrid and Piotr!

Roberts and Speed elected to the Fellowship of the Royal Society

I just found out that Gareth Roberts and Terry Speed have been elected as Fellows of the Royal Society (FRS). Congratulations to both for this prestigious recognition of their major contributions to Science! (Another Fellow elected this year is Bill Bryson, in recognition of his scientific popularisation books. Including one on the Royal Society I reviewed for CHANCE a few months ago.)