PhD thesis of Christian Schäfer

Today, I took part in the thesis committee of Christian Schäfer, a PhD student of Nicolas Chopin (hence a grand-child in this sense). The defence was both successful and interesting, focussing on proposal distributions for simulating binary vectors (with obvious applications to Bayesian model choice). The papers corresponding to the thesis are available on arXiv as

  • On parametric families for sampling binary data with specified mean and correlation arXiv:1111.0576
  • Particle algorithms for optimization on binary spaces arXiv:1111.0574
  • Sequential Monte Carlo on large binary sampling spaces (joint with Nicolas) arXiv:1101.6037
  • Parametric families on large binary spaces arXiv:1008.0055

While I had already mentioned the paper with Nicolas in an earlier post, I had not perceived the whole range of Christian’s work until today and I am quite impressed both by the methodological construct of sequential Monte Carlo in such unusual spaces and by the overall analysis of binary problems. While I think decision theory—i.e. the assessment of the impact of the choice of a particular subset of covariates—should play a larger role in the determination of relevant covariates in regression models, obtaining precise and robust estimates of the posterior probabilities of inclusion is a valuable feat.

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 )

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.

%d bloggers like this: