Blog Archives

Vectorize!

February 20, 2011
By
Vectorize!

Here is an email sent by one of my students a few days ago: Do you know how to integrate a function with an  “if”? For instance: >X=rnorm(100) >Femp=function(x){ +   return(sum(X<x)) +} >integrate(Femp,0,1)$value does not work. My reply was that the fundamental reason it does not work is that integrate (or curve for instance)

Read more »

UseR! 2011 in Warwick

February 20, 2011
By
UseR! 2011 in Warwick

This year useR! conference will take place in Warwick, on August 16-18.  It is being organised by the department of Statistics and funded by CRiSM and Revolution Analytics (providers of the R tee-shirt!). I wish I could attend but mid-August is usually associated with genuine (post-JSM) family vacations. Filed under: R, Statistics, University life Tagged:

Read more »

Parallel computation [permutations]

February 19, 2011
By
Parallel computation [permutations]

François Perron is visiting me for two months from Montréal and, following a discussion about the parallel implementation of MCMC algorithms—to which he also contributed with Yves Atchadé in 2005—, he remarked that a deterministic choice of permutations with the maximal contrast should do better than random or even half-random permutations. Assuming p processors or

Read more »

Le Monde puzzle [#6]

February 17, 2011
By
Le Monde puzzle [#6]

A simple challenge in Le Monde this week: find the group of four primes such that any sum of three terms in the group is prime and the overall sum is minimised. Here is a quick exploration by simulation, using the schoolmath package (with its imperfections): A=primes(start=1,end=53) lengthA=length(A) res=4*53 for (t in 1:10^4){ B=sample(A,4,prob=1/(1:lengthA)) sto=is.prim(sum(B))

Read more »

ABC in London

February 15, 2011
By
ABC in London

After the very exciting and I think quite successful ABC in Paris meeting two years ago, Michael Stumpf from Imperial College London suggested a second edition in London along the same lines. Michael kindly associated me with the planning of this meeting. It is (logically) called ABC in London (or ABCiL) and will take place

Read more »

Reaching 1000

February 14, 2011
By
Reaching 1000

This is the 1000th post on the ‘Og! Here are the entries that have had above 1000 views (not viewers) so far: In{s}a(ne)!! 5,353 “simply start over and build something better” 4,345 Julien on R shortcomings 1,966 Sudoku via simulated annealing 1,762 Of black swans and bleak prospects 1,462 Do we need an integrated Bayesian/likelihood

Read more »

Another Bernoulli factory

February 13, 2011
By
Another Bernoulli factory

The paper “Exact sampling for intractable probability distributions via a Bernoulli factory” by James Flegal and Radu Herbei got posted on arXiv without me noticing, presumably because it came out just between Larry Brown’s conference in Philadelphia and my skiing vacations! I became aware of it only yesterday and find it quite interesting in that

Read more »

Parallel computation [back]

February 12, 2011
By
Parallel computation [back]

We have now received reports back from JCGS for our parallel MCMC paper and they all are very nice and supportive! The reviewers essentially all like the Rao-Blackwellisation concept we developed in the paper and ask for additions towards a more concrete feeling for the practical consequences of the method. We should thus be able

Read more »

Le Monde puzzle [#5]

February 10, 2011
By
Le Monde puzzle [#5]

Another Sudoku-like puzzle from the weekend edition of Le Monde. The object it starts with is a 9×9 table where each entry is an integer and where neighbours take adjacent values. (Neighbours are defined as north, west, south and east of an entry.) The question is about whether or not it is possible to find

Read more »

Model weights for model choice

February 9, 2011
By
Model weights for model choice

An ‘Og reader. Emmanuel Charpentier, sent me the following email about model choice: I read with great interest your critique of Peter Congdon’s 2006 paper (CSDA, 50(2):346-357) proposing a method of estimation of posterior model probabilities based on improper distributions for parameters not present in the model inder examination, as well as a more general

Read more »