WSC 2[0]11
I have now registered for the WSC 2011 conference and I am looking forward the first day of talks tomorrow. Especially since, reading from the abstracts to the talks, it sounds as if many participants have a different understanding of the word simulation than I have. (I had the same impression this summer when taking part in a half-day of talks in Lancaster.) I am however slightly worried at having prepared my (advanced) tutorial for the right crowd, being unable to judge the background of the audience. Some of the talks are highly technical, others seem much more elementary… (I spent the whole night and morning, except for a fairly long and great run in the hills at sunrise, collating and adapting my slides from my graduate course and from different talks. The outcome is on slideshare.)
March 25, 2012 at 12:14 am
[…] Simulation Conference (WSC) that will take place in Berlin, December 9-12, this year, following Phoenix last year. I am organising a session there on statistical Monte Carlo methods with Nial Friel, […]
December 29, 2011 at 12:14 am
[…] it right away, going over the first chapter prior to giving it back. Later, on a plane trip between Phoenix and Minneapolis, I happened to sit next to a professional magician, The Amazing Hondo!, who started […]
December 13, 2011 at 11:36 am
[…] own session was not terribly well attended and, judging from some questions I got at the end I am still unsure […]
December 12, 2011 at 1:46 am
I’m curious about using the 1D stochastic volatility model as a motivation for ABC. Surely all of the computations with this model are O(T), which is the same cost as simulating it. In particular, without actually looking up the details, the manifold method of Girolami and Calderhead should only cost O(T) at each step. SImilar with using the INLA approximation as an independence sampler. For this to be an issue, T would need to be gargantuan, not just large. At which point I’m not sure ABC is particularly efficient (in terms of the number of samples required vs theoretically guaranteed quality of estimator).
December 12, 2011 at 10:57 am
Dan: I have always had trouble to make my MCMC work for this model, so it remains for me a challenging issue and an example of an unvailable w/o completion likelihood. (I agree there are ways to deal with it using those recent advance and I never tried ABC on the model.)