R finals

On the morning I returned from Varanasi and the ISBA meeting there, I had to give my R final exam (along with three of my colleagues in Paris-Dauphine). This year, the R course was completely in English, exam included, which means I can post it here as it may attract more interest than the French examens of past years…

I just completed grading my 32 copies, all from exam A, which takes a while as I have to check (and sometimes recover) the R code, and often to correct the obvious mistakes to see if the deeper understanding of the concepts is there. This year student cohort is surprisingly homogeneous: I did not spot any of the horrors I may have mentioned in previous posts.

I must alas acknowledge a grievous typo in the version of Exam B that was used the day of the final: cutting-and-pasting from A to B, I forgot to change the parameters in Exercise 2, asking them to simulate a Gamma(0,1). It is only after half an hour that a bright student pointed out the impossibility… We had tested the exams prior to printing them but this somehow escaped the four of us!

Now, as I was entering my grades into the global spreadsheet, I noticed a perfect… lack of correlation between those and the grades at the midterm exam. I wonder what that means: I could be grading at random, the levels in November and in January could be uncorrelated, some students could have cheated in November and others in January, student’s names or file names got mixed up, …? A rather surprising outcome!

estimating the measure and hence the constant

As mentioned on my post about the final day of the ICERM workshop, Xiao-Li Meng addresses this issue of “estimating the constant” in his talk. It is even his central theme. Here are his (2011) slides as he sent them to me (with permission to post them!):

He therefore points out in slide #5 why the likelihood cannot be expressed in terms of the normalising constant because this is not a free parameter. Right! His explanation for the approximation of the unknown constant is then to replace the known but intractable dominating measure—in the sense that it cannot compute the integral—with a discrete (or non-parametric) measure supported by the sample. Because the measure is defined up to a constant, this leads to sample weights being proportional to the inverse density. Of course, this representation of the problem is open to criticism: why focus only on measures supported by the sample? The fact that it is the MLE is used as an argument in Xiao-Li’s talk, but this can alternatively be seen as a drawback: I remember reviewing Dankmar Böhning’s Computer-Assisted Analysis of Mixtures and being horrified when discovering this feature! I am currently more agnostic since this appears as an alternative version of empirical likelihood. There are still questions about the measure estimation principle: for instance, when handling several samples from several distributions, why should they all contribute to a single estimate of μ rather than to a product of measures? (Maybe because their models are all dominated by the same measure μ.) Now, getting back to my earlier remark, and as a possible answer to Larry’s quesiton, there could well be a Bayesian version of the above, avoiding the rough empirical likelihood via Gaussian or Drichlet process prior modelling.

ICERM, Brown, Providence, RI (#3)

Just yet another perfect day in Providence! Especially when I thought it was going to be a half-day: After a longer and slightly warmer run in the early morning around the peninsula, I attended the lecture by Eric Moulines on his recent results on adaptive MCMC and the equi-energy sampler. At this point, we were told that, since Peter Glynn was sick, the afternoon talks were drifted forward. This meant that I could attend Mylène Bédard’s talk in the morning and most of Xiao-Li Meng’s talk, before catching my bus to the airport, making it a full day in the end!

The research presented by Mylène (and coauthored with Randal Douc and Eric Moulines) was on multiple-try MCMC and delayed-rejection MCMC, with optimal scaling results and a comparison of the efficiency of those involved schemes. I had not seen the work before and got quite impressed by the precision of the results and the potential for huge efficiency gains. One of the most interesting tricks was to use an antithetic move for the second step, considerably improving the acceptance rate in the process. An aside exciting point was to realise that the hit-and-run solution was also open to wide time-savings thanks to some factorisation.

While Xiao-Li’s talk had connections with his earlier illuminating talk in New York last year, I am quite desolate to have missed [the most novel] half of it (and still caught my bus by a two minute margin!), esp. because it connected beautifully with the constant estimation controverse! Indeed, Xiao-Li started his presentation with the pseudo-paradox that the likelihood cannot be written as a function of the normalising constant, simply because this is not a free parameter. He then switched to his usual theme that the dominating measure was to be replaced with a substitute and estimated.The normalising constant being a function of the dominating measure, it is a by-product of this estimation step. And can even be endowed within a Bayesian framework. Obviously, one can always argue against the fact that the dominating measure is truly unknown, however this gives a very elegant safe-conduct to escape the debate about the constant that did not want to be estimated…So to answer Xiao-Li’s question as I was leaving the conference room, I have now come to a more complete agreement with his approach. And think further advances could be contemplated along this path…

workshop a Venezia (2)

I could only attend one day of the workshop on likelihood, approximate likelihood and nonparametric statistical techniques with some applications, and I wish I could have stayed a day longer (and definitely not only for the pleasure of being in Venezia!) Yesterday, Bruce Lindsay started the day with an extended review of composite likelihood, followed by recent applications of composite likelihood to clustering (I was completely unaware he had worked on the topic in the 80′s!). His talk was followed by several talks working on composite likelihood and other pseudo-likelihoods, which made me think about potential applications to ABC. During my tutorial talk on ABC, I got interesting questions on multiple testing and how to combine the different “optimal” summary statistics (answer: take all of them, it would not make sense to co;pare one pair with one summary statistic and another pair with another summary statistic), and on why we were using empirical likelihood rather than another pseudo-likelihood (answer: I do not have a definite answer. I guess it depends on the ease with which the pseudo-likelihood is derived and what we do with it. I would e.g. feel less confident to use the pairwise composite as a substitute likelihood rather than as the basis for a score function.) In the final afternoon, Monica Musio presented her joint work with Phil Dawid on score functions and their connection with pseudo-likelihood and estimating equations (another possible opening for ABC), mentioning a score family developped by Hyvärinen that involves the gradient of the square-root of a density, in the best James-Stein tradition! (Plus an approach bypassing the annoying missing normalising constant.) Then, based on a joint work with Nicola Satrori and Laura Ventura, Ruli Erlis exposed a 3rd-order tail approximation towards a (marginal) posterior simulation called HOTA. As Ruli will visit me in Paris in the coming weeks, I hope I can explore the possibilities of this method when he is (t)here. At last, Stéfano Cabras discussed higher-order approximations for Bayesian point-null hypotheses (jointly with Walter Racugno and Laura Ventura), mentioning the Pereira and Stern (so special) loss function mentioned in my post on Måns’ paper the very same day! It was thus a very informative and beneficial day for me, furthermore spent in a room overlooking the Canal Grande in the most superb location!

ultimate R recursion

One of my students wrote the following code for his R exam, trying to do accept-reject simulation (of a Rayleigh distribution) and constant approximation at the same time:

fAR1=function(n){
 u=runif(n)
 x=rexp(n)
 f=(C*(x)*exp(-2*x^2/3))
 g=dexp(n,1)
 test=(u<f/(3*g))
 y=x[test]
 p=length(y)/n #acceptance probability
 M=1/p
 C=M/3
 hist(y,20,freq=FALSE)
 return(x)
 }

which I find remarkable if alas doomed to fail! I wonder if there exists a (real as opposed to fantasy) computer language where you could introduce constants C and only define them later… (What’s rather sad is that I keep insisting on the fact that accept-reject does not need the constant C to operate. And that I found the same mistake in several of the students’ code. There is a further mistake in the above code when defining g. I also wonder where the 3 came from…)