Dutch summer workshops on Bayesian modeling

Just received an email about two Bayesian workshops in Amsterdam this summer:

• “Theory and Practice of Bayesian Hypothesis Testing, A JASP Workshop”  August 22 – August 23, 2019
• “Bayesian Modeling for Cognitive Science, A JAGS and WinBUGS Workshop” August 26 – August 30

both taking place at the University of Amsterdam. And focussed on Bayesian software.

absolutely no Bayesians inside!

ABCπ

Ritabrata Dutta, Marcel Schöengens, Jukka-Pekka Onnela, and Antonietta Mira recently put a new ABC software on-line, called ABCpy for ABC with Python. The software aims at  an automated parallelisation of ABC runs, requiring only code to generate from the (generative) model and the choice of summary statistics and of associated distance. Alternatively an approximate likelihood (as in synthetic likelihood) can be used. The tolerance ε is chosen as a percentile of the prior predictive distribution on the distance. The versions of ABC found in ABCpy are

1. Population Monte Carlo for ABC (PMCABC);
2. sequential Monte Carlo ABC (ABC-SMC);
3. replenishment Sequential Monte Carlo ABC (RSMC-ABC);
4. adaptive Population Monte Carlo ABC (APMCABC);
5. ABC with subset simulation (ABCsubsim); and
6. simulated annealing ABC (SABC)

Anto mentioned ABCpy to me while in Harvard last week and I have not tested the program (my only brush with Python being the occasional call to latex2wp for SeriesB’log). And obviously, writing a blog about Monte (Carlo and) Python makes a link to the Monty Pythons irresistible:

STAN trailer [PG+53]

[Heading off to mountainous areas with no Internet or phone connection, I posted a series of entries for the following week, starting with this brilliant trailer of Michael:]

MCMSki IV, Jan. 6-8, 2014, Chamonix (news #8)

A reposted item of news about MCMSki IV: as posted by Brad Carlin this afternoon to the Biometrics Section and Bayesian Statistical Science Section of the ASA,

The fifth joint international meeting of the IMS (Institute of Mathematical Statistics) and ISBA (International Society for Bayesian Analysis), nicknamed “MCMSki IV”, will be held in Chamonix Mont-Blanc, France, from Monday, January 6 to Wednesday, January 8, 2014.  The meeting, the first for the newly-created BayesComp section of ISBA, will focus on all aspects of MCMC theory and methodology, including related fields like sequential Monte Carlo, approximate Bayesian computation, Hamiltonian Monte Carlo.  In contrast with the earlier meetings, it will merge the satellite Adap’ski workshop into the main meeting by having parallel invited and contributed sessions on those different themes, as well as poster sessions on both Monday and Tuesday nights.  In addition, a one-day post-conference satellite workshop on Bayesian nonparametrics, modelling and computations (“BNPski)” will be held in the same location on January 9th, 2014.

Please see our conference website,

http://www.pages.drexel.edu/~mwl25/mcmski/

for more information, including links to the preliminary program, lodging and travel information, and our conference registration page.  Please note the “early bird” registration deadline of *October 15, 2013*.

Finally (and most importantly for some), we are very pleased to announce that we have received funds from ISBA, SBSS, and other sources sufficient to help support the travel expenses of some junior investigators (defined as current PhD student, or less than 5 years since PhD). Information about how to apply for this support can be found at:

http://www.pages.drexel.edu/~mwl25/mcmski/student.html

Note that to apply you must first register with the ISBA website (if you have not done so already) and submit your abstract to the ISBA Abstract Page.  Full directions are given at the link above.  The deadline to apply for this financial support is also October 15, 2013.

We look forward to welcoming you in Chamonix this January!

Best,

Brad Carlin, Antonietta Mira, and Christian Robert
MCMSki IV conference co-organizers