Bayesian News Feeds
Filed under: University life Tagged: oolong, shui xian, tea, 水仙
As the cold I caught in England (maybe) last week was getting more severe, I went to see my doctor yesterday and he diagnosed a flu… So expect delays on the ‘Og in the coming days. (Interestingly, I have reached the part in Nate Silver’s book, The Signal and the Noise, about epidemics and the poor predictions related to the swine flu a few years ago.) What about Bach?! Well, my doctor always has classical music in his waiting room and Bach’s 6th Brandenburg concerto was playing while I was waiting, sneezing, and coughing. This piece is associated with my early childhood as it was on the only vinyl record my parents owned for quite a while, so we ended up listening to it very regularly… (In case you wonder, I still enjoy it very much!)
Filed under: Uncategorized Tagged: Bach, Brandenbug concertos, Nate Silver, The Signal and The Noise
This recent arXiv posting by Martin Weinberg and co-authors was pointed out to me by friends because of its title! It indeed sounded a bit inflated. And also reminded me of old style papers where the title was somehow the abstract. Like An Essay towards Solving a Problem in the Doctrine of Chances… So I had a look at it on my way to Gainesville. The paper starts from the earlier paper by Weinberg (2012) in Bayesian Analysis where he uses an HPD region to determine the Bayes factor by a safe harmonic mean estimator (an idea we already advocated earlier with Jean-Michel Marin in the San Antonio volume and with Darren Wraith in the MaxEnt volume). An extra idea is to try to optimise [against the variance of the resulting evidence] the region over which the integration is performed: “choose a domain that results in the most accurate integral with the smallest number of samples” (p.3). The authors proceed by volume peeling, using some quadrature formula for the posterior coverage of the region, either by Riemann or Lebesgue approximations (p.5). I was fairly lost at this stage and the third proposal based on adaptively managing hyperrectangles (p.7) went completely over my head! The sentence “the results are clearly worse with O(∞) errors, but are still remarkably better for high dimensionality”(p.11) did not make sense either… The method may thus be remarkably simple, but the paper is not written in a way that conveys this impression!
Filed under: Statistics Tagged: Bayes factors, harmonic mean, HPD region, Lebesgue integration, MaxEnt2009, quadrature, Riemann integration, San Antonio, Thomas Bayes
The current puzzle is as follows:
All integers are associated with one of two colours (gold and silver, say!), with the rule that for any triplet (with possible replications) with the same colour, the sum of this triplet is also of the same colour. If 58 is golden and a lot of integers are silver, what is the colour of 21? and of 40?
Here is my resolution (spoiler alert!). The above rule tells us that for any x of a given colour, all multiples of the form 3px are also of that colour. Also, given that 3=1+1+1, 5=3+1+1, &tc., all odd numbers are of the same colour. Including 21. Starting with 2, a similar reasoning tells us that all integers of the form 2+4n are of the same colour as 2, which includes 58. Hence 2 is golden. Now, were 1 golden as well, 4=2+1+1 would be golden and all integers would be golden. Therefore, 1 is silver, therefore 21 is silver. Next, considering 4, if it was silver, 6=4+1+1 would be silver as well: impossible! Therefore, 4 is golden, which implies that all integers of the form 4+4n are also golden, hence 40 is golden. (Actually, all even integers are golden and all odd integers are silver.) The R resolution of this puzzle is not so obvious: either one sorts out that only 1,2, and 4 matter, in which case the issue is resolved, or else one tries “at random” but this does not mean much in this case: should one check 2⁵⁸ colourings? build a tree algorithm that severs impossible branches?
Filed under: Books, Kids Tagged: Le Monde, mathematical puzzle, R
Another focussed spam in the mail:
Dear Dr. Christian P. Robert,
How are you?
I read your interesting article of “Error and inference: an outsider stand on a frequentist philosophy“, and I know you are an active professional in this field.
Now, I am writing you to call for new papers, on behalf of Review of Economics & Finance, which is an English quarterly journal in Canada.
This journal is currently indexed by EconLit of American Economic Association (AEA), EBSCO, RePEc, National Bibliography of Canada, Library and Archives Canada, DOAJ, Ulrich, and so on.
The publication fee is CAD$450, if your paper is qualified for publication after refereeing. The submission fee of $50 is NOT applied to you by March 6th, 2013.
Thank you for your consideration. Have a rewarding month!
At least, they are quite honest about the cost of publishing there. But they should check on which paper they pick rather than using a robot that takes a paper not talking about economics or finance… (And why on Earth this line about the rewarding month?!)
Filed under: Statistics, University life Tagged: Canada, commercial editor, Error and Inference, Review of Economics & Finance, spam, Theory and Decisions
- Andrew Gelman on Introducing Monte Carlo Methods with R
- Bill Strawderman on Statistical Inference
- Jean-Louis Foulley on Variance Components
- Larry Wasserman on Theory of Point Estimation
- Xiao-Li Meng on Monte Carlo Statistical Methods
Although all of those books have appeared between twenty and five years ago, the reviews are definitely worth reading! (Disclaimer: I am the editor of the Books Review section who contacted friends of George to write the reviews, as well as the co-author of two of those books!) They bring in my (therefore biased) opinion a worthy evaluation of the depths and impacts of those major books, and they also reveal why George was a great teacher, bringing much into the classroom and to his students… (Unless I am confused the whole series of reviews is available to all, and not only to CHANCE subscribers. Thanks, Sam!)
Filed under: Books, R, Statistics, University life Tagged: CHANCE, George Casella, Introducing Monte Carlo Methods with R, Monte Carlo Statistical Methods, Sam Behesta, statistical inference, Theory of Point Estimation, Variance Components
Just read a puzzling fact in Le Monde science leaflet: someone at the “Great Internet Mersenne Prime Search” (GIMPS) just found a prime number with 17 million digits and…it is the 48th prime number of the form 2p-1. Which just means it is an odd number. How odd a remark!!! And how wrong. In fact, the short news item meant that this is a Mersenne number, of the form 2p-1! Another victim of a hasty cut&paste, I presume…
Filed under: Books Tagged: cut and paste, Le Monde, Mersenne numbers, prime numbers, the Great Internet Mersenne Prime Search
Deliverance!!! We have at last completed our book! Bayesian Essentials with R is off my desk! In a final nitty-gritty day of compiling and recompiling the R package bayess and the LaTeX file, we have reached versions that were in par with our expectations. The package has been submitted to CRAN (it has gone back and forth a few times, with requests to lower the computing time in the examples: each example should take less than 10s, then 5s…), then accepted by CRAN, incl. a Windows version, and the book has be sent to Springer-Verlag. This truly is a deliverance for me as this book project has been on my work horizon almost constantly for more than the past two years, led to exciting times in Luminy, Carnon and Berlin, has taken an heavy toll on my collaborations and research activities, and was slowly turning into a unsavoury chore! I am thus delighted Jean-Michel and I managed to close the door before any disastrous consequence on either the book or our friendship could develop. Bayesian Essentials with R is certainly an improvement compared with Bayesian Core, primarily by providing a direct access to the R code. We dearly hope it will attract a wider readership by reducing the mathematical requirements (even though some parts are still too involved for most undergraduates) and we will keep testing it with our own students in Montpellier and Paris over the coming months. In the meanwhile, I just enjoy this feeling of renewed freedom!!!
Filed under: Books, pictures, R, Statistics Tagged: Bayesian Core, Bayesian Essentials with R, bayess, Berlin, books, Carnon, CRAN, La Défense, Luminy, R, Springer-Verlag
Filed under: Books, pictures, Travel Tagged: books, England, Euston, London, Robert Stevenson, Scotland
Today I took the Eurostar to London to give a seminar at the Gatsby Computational Neuroscience Unit, UCL. (Just a few blocks from the London School of Hygiene and Tropical Diseases, where I gave an ABC talk last year.) I had great fun, thanks to an uninterrupted sequence of meetings: I got a crash course on RKHS (reproducible kernel Hilbert spaces) by Arthur Gretton, discussed about estimating the number of species, dealing with unknown functions of the parameter in the likelihood, using tests as ABC statistics, and explained how to use empirical likelihoods in non-iid settings. After this full day, we had a superb dinner at St. John, a Michelin starred restaurant with highly enjoyable English cuisine, offering game and offal dishes that reminded me of Le Petit Marguery in Paris… (Not a place for vegetarians, obviously.)
Filed under: Statistics, Travel, University life Tagged: ABC, empirical likelihood, Eurostar, Gatsby, London, RKHS, seminar, St. John restaurant, UCL