MCMC in Warwickshire (end)

I am back from the EPSRC Symposium Workshop on Markov Chain-Monte Carlo organised at the University of Warwick, by Elke Thonnes and Gareth Roberts. There were mostly exciting talks yesterday, once again, out of which I want to single out

  1. the extension by Neal Friel of our knn paper with Lionel Cucala, Jean-Michel Marin and Mike Titterington (also available on arXiv and soon to appear in Jasa). The extension replaces the binary aspect of the original knn model with a distance weighted penalty;
  2. an interesting analysis of the CMB model by Simon Wilson, using multilevel mixtures to decompose the sources of noise, even though the analysis cannot apply so far to the true data;
  3. a very clear defence of the use of fixed width analysis of MCMC output by Galin Jones that had me convinced it should be included in the R book (Chapter 8);
  4. another crisp presentation by Ajay Jasra of non-linear (adaptive) MCMC algorithms that allow for random restarts in the past of the chain while preserving the convergence.

While I missed the talks today because of an early flight back, and I regret it, I nonetheless come back from the meeting full of energy and with new ideas for papers and research directions, both for me and my (future) students. This does not happen with every meeting, especially with generic ones, so this is for me a very very successful workshop for which I congratulate the organisers.

5 Responses to “MCMC in Warwickshire (end)”

  1. […] Flegal posted on arXiv a new version of their batch mean convergence assessment paper. As posted earlier, I think this is a very interesting approach to MCMC convergence assessment and stopping rule, and […]

  2. […] version, written with Randal Douc and now posted on arXiv, we found—during this inspiring MCMC meeting at the University of Warwick—a different representation of the Metropolis-Hastings estimators […]

  3. tu nous as pas parle de la qualite de l’accueil.. de la biere???
    cela en ferait un vrai compte rendu de congres…
    plus serieusement, j’ai lu le papier de Galin Jones… et, je ne k’ai pas trouve si convaincant… je regarderais ca a nouveau…

    • Merci arno pour ces commentaires toujours aussi profonds!!! En ce qui concerne l’exposé de Galin, j’ai bien aimé la clareté des conditions pour appliquer des batch-means afin de controler la convergence d’un mcmc, sans attendre l’asymptotique. Sans doute à rajouter au “MCMC with R”…

  4. […] Xi’an’s Og An attempt at bloggin, from scratch… « MCMC in Warwickshire (end) […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.