Archive for Eiffel Peak

posterior distribution missing the MLE

Posted in Books, Kids, pictures, Statistics with tags , , , , , , , on April 25, 2019 by xi'an

An X validated question as to why the MLE is not necessarily (well) covered by a posterior distribution. Even for a flat prior… Which in restrospect highlights the fact that the MLE (and the MAP) are invasive species in a Bayesian ecosystem. Since they do not account for the dominating measure. And hence do not fare well under reparameterisation. (As a very much to the side comment, I also managed to write an almost identical and simultaneous answer to the first answer to the question.)

Off to Banff!!

Posted in Books, Mountains, R, Statistics, Travel, University life with tags , , , , , , , , , on September 10, 2010 by xi'an

Today I am travelling from Paris to Banff, via Amsterdam and Calgary, to take part in the Hierarchical Bayesian Methods in Ecology two day workshop organised at BIRS by Devin Goodsman (University of Alberta),  François Teste (University of Alberta), and myself. I am very excited both by the opportunity to meet young researchers in ecology and forestry, and by the prospect in spending a few days in the Rockies, hopefully with an opportunity to go hiking, scrambling and even climbing. (Plus the purely random crossing of Julien‘s trip in this area!) The slides will be mostly following those of the course I gave in Aosta, while using Introducing Monte Carlo Methods with R for R practicals:

Bayes in the Rockies

Posted in Books, Mountains, Statistics, University life with tags , , , , , , , , , , on June 25, 2010 by xi'an

A few months ago, following the course I gave last summer in Vale dAosta, I got contacted by Devin Goodsman from the University of Alberta, in order to organise a short course there for environmental scientists in hierarchical Bayes modelling and its R implementation. I (obviously) accepted the invitation and, thanks to Devin and François Teste, we managed to get the short course housed at the wonderful BIRS (Banff International Research Station for Mathematical Innovation and Discovery) centre in Banff, in the very heart of the Canadian Rockies. Last time I was at BIRS, it was four year ago for the 07w5079 Bioinformatics, Genetics and Stochastic Computation: Bridging the Gap workshop, organised jointly with Arnaud Doucet and Raphaël Gottardo from UBC, and I enjoyed the place and organisation immensely. (Not to mention the surroundings, as shown this picture of Mount Temple, climbed in 2002, taken from Eiffel Peak.)

The current workshop, Hierarchical Bayesian Methods in Ecology, is numbered 10w2170 and will take place over the weekend of September 10-12 this Fall.  The schedule is provided on the workshop webpage. I plan to cover Chapters 2, 3 and 4 of Bayesian Core on Saturday, before spending all of Sunday morning and early afternoon on hierarchical models. I actually plan to take advantage of the revision of Bayesian Core in Luminy next week to include a chapter on hierarchical modelling. Now, there are still a few places left in order to reach the 20 participant (strict) upper limit, so, if you are interested in attending the workshop/course, contact me at my gmail email account: bayesianstatistics. (The cost is $220, which includes two nights of lodging at BIRS and a copy of Introducing Monte Carlo Methods with R! Meals and transportation are not included.)