meandering

A bit of a misunderstanding from Randall Munroe and then some: the function F returns a triplet, hence G should return a triplet as well. Even if the limit does return three identical values. And he should have also included the (infamous) harmonic mean! And the subtext (behind the picture) mentions random forest statistics, using every mean one can think of and dropping those that are doing worse, while here all solutions return the same value, hence do not directly discriminate between the averages (and there is no objective function to create the nodes in the trees, &tc.).

Here is a test R code including the harmonic mean:

xkcd=function(x)c(mean(x),exp(mean(log(x))),median(x),1/mean(1/x))
xxxkcd=function(x,N=10)ifelse(rep(N==1,4),xkcd(x),xxxkcd(xkcd(x),N-1))
xxxkcd(rexp(11))
[1] 1.018197 1.018197 1.018197 1.018197

locusts in a random forest

My friends from Montpellier, where I am visiting today, Arnaud Estoup, Jean-Michel Marin, and Louis Raynal, along with their co-authors, have recently posted on biorXiv a paper using ABC-RF (Random Forests) to analyse the divergence of two populations of desert locusts in Africa. (I actually first heard of their paper by an unsolicited email from one of these self-declared research aggregates.)

“…the present study is the first one using recently developed ABC-RF algorithms to carry out inferences about both scenario choice and parameter estimation, on a real multi-locus microsatellite dataset. It includes and illustrates three novelties in statistical analyses (…): model grouping analyses based on several key evolutionary events, assessment of the quality of predictions to evaluate the robustness of our inferences, and incorporation of previous information on the mutational setting of the used microsatellite markers”.

The construction of the competing models (or scenarios) is built upon data of past precipitations and desert evolution spanning several interglacial periods, back to the middle Pleistocene, concluding at a probable separation in the middle-late stages of the Holocene, which corresponds to the last transition from humid to arid conditions in the African continent. The probability of choosing the wrong model is exploited to determine which model(s) lead(s) to a posterior [ABC] probability lower than the corresponding prior probability, and only one scenario stands this test. As in previous ABC-RF implementations, the summary statistics are complemented by pure noise statistics in order to determine a barrier in the collection of statistics, even though those just above the noise elements (which often cluster together) may achieve better Gini importance by mere chance. An aspect of the paper that I particularly like is the discussion of the various prior modellings one can derive from existing information (or lack thereof) and the evaluation of the impact of these modellings on the resulting inference based on simulated pseudo-data.

ABC in Les Diablerets

Since I could not download the slides of my ABC course in Les Diablerets in one go, I broke them by chapters as follows. (Warning: there is very little novelty in those slides, except for the final part on consistency.)

Although I did not do it on purpose (!), starting with indirect inference and other methods inspired from econometrics induced some discussion in the first hour of the course with econometricians in the room. Including Elvezio Ronchetti.

I also regretted piling too much material in the alphabet soup, as it was too widespread for a new audience. And as I could not keep the coherence of the earlier parts by going thru so many papers at once. Especially since I was a bit knackered after a day of skiing….

I managed to get to the final convergence chapter on the last day, even though I had to skip some of the earlier material. Which should be reorganised anyway as the parts between model choice with random forests and inference with random forests are not fully connected!

JSM 2014, Boston [#3]

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.