Archive for April, 2012

mad statistic

Posted in R, Statistics, University life with tags , , , , , on April 30, 2012 by xi'an

In the motivating toy example to our ABC model choice paper, we compare summary statistics, mean, median, variance, and… median absolute deviation (mad). The latest is the only one able to discriminate between our normal and Laplace models (as now discussed on Cross Validated!). When rerunning simulations to produce nicer graphical outcomes (for the revision), I noticed a much longer run time associated with the computation of the mad statistic. Here is a comparison for the computation of the mean, median, and mad on identical simulations:

> system.time(mmean(10^5))
   user  system elapsed
  4.040   0.056   4.350
> system.time(mmedian(10^5))
user  system elapsed
12.509   0.012  15.353
> system.time(mmad(10^5))
   user  system elapsed
 23.345   0.036  23.458

Now, this is not particularly surprising: computing a median takes longer than computing a mean, even using quicksort!, hence computing two medians… Still, having to wait about six times longer for the delivery of a mad statistics is somehow…mad!

art brut

Posted in pictures, Running with tags , , on April 29, 2012 by xi'an

The Redeemer (Jo Nesbo)

Posted in Books, Travel with tags , , , , , , on April 28, 2012 by xi'an

I picked this book in Oxford two months ago with some reticence because of “The next Stieg Larsson” sticker on it… Indeed, I did not like the underlying message of the Larsson Millenium trilogy, even though I admired the efficiency of the story-telling. Now, The Redeemer is the first book by Jo Nesbo I read and I rather liked it, at least conditional on the serial killer genre. Maybe the fact that it takes place in Oslo, a city I particularly like, makes it more interesting. Maybe the convoluted psychological features of the detective Harry and of the killers are much more convincing than in Larsson‘s books.

And our prejudices solve cases. Because they are not based on lack of knowledge, but on actual facts and experience. In this room we reserve the right to discriminate against everyone, regardless of race, religion, or gender. Our defence is that it is not exclusively the weakest members of the society  who are discriminated against (…) Since we work with probabilities and limited knowledge, we cannot afford to ignore knowledge wherever we find it.” J. Nesbo, The Redeemer (p. 143)

The central character is the detective, Harry Hole, who is looking as much for his true self than for the murderer. He is fighting against alcoholism, which almost had him thrown out of the police, against religious fanaticisms, against corruption within the force, against turning sexual encounters into longer term relationships and against regrets about his separation from his girlfriend Rakel, but (minor spoiler!) falls short of winning all those battles. Other characters are also well-built, from the professional assassin to the highly various actors from the Salvation Army. And the underlying theme of young girls’ abuses make the quest for the assassin more dramatic, with the endings completely unexpected. (If somewhat unrealistic.) I also like the understated way the story unfolds, which sounds very suited to snow-encased Oslo (even though some of its harsher aspects emerge at times). I should have read the three previous novels by Jo Nesbo in the series, but The Redeemer can easily be read as a stand-alone. Not perfect, but quite enjoyable and definitely gripping.

camera miracles: once, not twice!

Posted in Books, Mountains, pictures, Travel with tags , , , , , , , , , , on April 27, 2012 by xi'an

As I mentioned in a post last February, I almost lost my (Nikon Coolpix L26) camera to the cloaca maxima, in Roma. It however remained (miraculously) within reach inside the manhole there… Well, this kind of miracle does not happen twice (or only in Roma…)  and I have now lost the camera for good! When climbing Tower Ridge, after the first belay to go up Douglas gap, I took it out of my pocket to take a few pictures of the beginning of the ridge and of the fantastic view of that side of Ben Nevis. As I was mostly paying attention to Kenny going up the blocks above us (to make sure of my holds there), I did not look as I put my camera back inside my overpants and it slid out of the pocket, swiftly accelerating down the snowy slopes to disappear into Coire na Ciste… There was no way we were going to check whether or not it was retrievable, so I called myself a few well-chosen names and we continued our climb along the ridge without further delay. In fact, I had another camera in my bag, my older and bulkier Konica Minolta Dimage Z20, but it was impossible to get hold of it in most places (as I would have had to unpack) and it anyway ran out of battery (which explains why I have so few pictures of the top of the Ben and of the unbelievable [and rare] views of the Highlands invading the ‘Og in the past days!).

Here is thus the last picture taken from my lost camera, a view of the Aonach Eagach ridge from the bottom of Glencoe (and the start of the trail to the Lost Valley). Apart from this miracle in Roma, I have been rather unlucky with cameras lately, loosing first my favourite one in a New York taxi, then this one on Tower Ridge. Actually, I consoled myself with the fact that the quality of this Nikon Coolpix L26 camera was rather unsatisfactory, behaving poorly in anything but clear weather and having grown a mark (fungus?) on the lens (after falling in the snow during my X’mas ski trip). Mark that is clearly visible on the right of  the ptarmigan picture below. Anyway, I will now have to look for a new camera, hopefully supported by ‘Og’s readers (!) via the links to Amazon.com and Amazon.fr there, which earn me a monetary gain [of 4% to 7%] if a purchase [of any product] is made within the 24 hours following the entry on Amazon through this link, thanks to the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to amazon.com/fr.

Edinburgh sunset

Posted in pictures, Travel, University life with tags , , , , , on April 26, 2012 by xi'an

MCMC at ICMS (3)

Posted in pictures, Statistics, Travel, University life with tags , , , , , , , , , , , , , , , on April 26, 2012 by xi'an

The intense pace of the two first days of our workshop on MCMC at ICMS had apparently taken an heavy toll on the participants as a part of the audience was missing this morning! Although not as a consequence of the haggis of the previous night at the conference dinner, nor even as a result of the above pace. In fact, the missing participants had opted ahead of time for leaving the workshop early, which is understandable given everyone’s busy schedule, esp. for those attending both Bristol and Edinburgh workshops, however slightly impacting the atmosphere of the final day. (Except for Mark Girolami who most unfortunately suffered such a teeth infection that he had to seek urgent medical assistance yesterday afternoon. Best wishes to Mark for a prompt recovery, say I with a dental appointment tomorrow…!)

The plenary talk of the day was delivered by Heikki Haario, who provided us with a survey of the (adaptive) MCMC advances he and his collaborators had made in the analysis of complex and immensely high-dimensional weather models. This group of Finnish researchers, who started from inverse problem analysis rather than from MCMC, have had a major impact on the design and validation of adaptive MCMC algorithms, especially in the late 1990’s. (Heikki also was a co-organizer of the Adap’ski workshops, workshops that may be continued, stay tuned!) The next talk, by Marko Laine, was also about adaptive MCMC algorithms, with the difference that the application was climate modelling. It contained interesting directions about early stopping (“early rejection”, as opposed to “delayed rejection”) of diverging proposals (gaining 80% in computing time!) and about parallel adaptation. Still in the same theme, Gersende Fort explained the adaptive version of the equi-energy sampler she and co-authors had recently developed. Although she had briefly presented this paper in Banff a month ago, I found the talk quite informative about the implementation of the method and at the perfect technical level (for me!).

In [what I now perceive as] another recurrent theme of the workshop, namely the recourse to Gaussian structures like Gaussian processes (see, e.g., Ian Murray’s talk yesterday), Andrew Stuart gave us a light introduction to random walk Metropolis-Hastings algorithms on Hilbert spaces. In particular, he related to Ian Murray’s talk of yesterday as to the definition of a “new” random walk (due to Radford Neal)  that makes a proposal

y=\sqrt{1-\beta^2}x_{t-1}+\beta\zeta\quad 0<\beta<1,\zeta\sim\varphi(|\zeta|)

that still preserves the acceptance probability of the original (“old”) random walk proposal. The final talks of the morning were Krys Latuszynski’s and Nick Whiteley’s very pedagogical presentations of the convergence properties of manifold MALA and of particle filters for hidden Markov models.  In both cases, the speakers avoided the overly technical details and provided clear intuition in the presented results, a great feat after those three intense days of talks! (Having attended Nick’s talk in Paris two weeks ago helped of course.)

Unfortunately, due to very limited flight options (after one week of traveling around the UK) and also being slightly worried at the idea of missing my flight!, I had to leave the meeting along with all my French colleagues right after Jean-Michel Marin’s talk on (hidden) Potts driven mixtures, explaining the computational difficulties in deriving marginal likelihoods. I thus missed the final talk of the workshop by Gareth Tribello. And delivering my final remarks at the lunch break.

Overall, when reflecting on those two Monte Carlo workshops, I feel I preferred the pace of the Bristol workshop, because it allowed for more interactions between the participants by scheduling less talks… This being said, the organization at ICMS was superb (as usual!) and the talks were uniformly very good so it also was a very profitable meeting, of a different kind! As written earlier, among other things, it induced (in me) some reflections on a possible new research topic with friends there. Looking forward to visit Scotland again, of course!

Follow

Get every new post delivered to your Inbox.

Join 634 other followers