Research

I’m an assistant professor in Statistics at the LUMC and the Mathematical Institute of Leiden University.

Office MI: 221, Snellius building.
Office LUMC: S-05-42, building 2.
E-mail: svdpas at math dot leidenuniv dot nl
LinkedIn

Research interests
Bayesian nonparametrics; Sparsity; Nonparametric regression; Community detection; Survival analysis; Competing risks.

Conference: Survival Analysis for Junior Researchers 2018
SAfJR 2018 will be hosted in Leiden! See the conference website for more information.

Publications
Google Scholar

Preprints

10. Posterior Concentration for Bayesian Regression Trees and their Ensembles. [pdf] (with Veronika Rocková)

9. Bayesian Dyadic Trees and Histograms for Regression. [pdf] (with Veronika Rocková, accepted for a poster presentation at NIPS 2017)

8. Bayesian Community Detection. [pdf] (with Aad van der Vaart, to appear in Bayesian Analysis)

7. Uncertainty Quantification for the Horseshoe. [pdf] (with Botond Szabó and Aad van der Vaart, to appear in Bayesian Analysis)

6. Almost the Best of Three Worlds: Risk, Consistency and Optional Stopping for the Switch Criterion in Nested Model Selection. [pdf] (with Peter Grünwald, to appear in Statistica Sinica)

Published

5. S. van der Pas, B. Szabó and A. van der Vaart (2017). Adaptive Posterior Contraction Rates for the Horseshoe. Electronic Journal of Statistics 11 (2), 3196-3225. [pdf]

4. S.L. van der Pas, R.G.H.H. Nelissen and M. Fiocco (2017). Patients with Staged Bilateral Total Joint Arthroplasty in Registries: Immortal Time Bias and Methodological Options. Journal of Bone & Joint Surgery – American Volume 99 (15), p e82. [pdfLink to online journal edition here. Copyright JBJS. 

3. S.L. van der Pas, J.-B. Salomond and J. Schmidt-Hieber (2016). Conditions for Posterior Contraction in the Sparse Normal Means Problem. Electronic Journal of Statistics 10, 976-1000. [pdf]

2. S.L. van der Pas, B.J.K Kleijn and A.W. van der Vaart (2014). The Horseshoe Estimator: Posterior Concentration around Nearly Black Vectors. Electronic Journal of Statistics 8 (2), 2585-2618. [pdf]

1. S. van der Pas (2014). The Normal Road to Geometry: in Euclid’s Elements and the Mathematical Competence of His Audience. The Classical Quarterly 64, 558-573. [pdf] Link to online journal edition here. Copyright The Classical Association. 

Software

The R package `horseshoe’ is available on CRAN here, or can be installed by typing install.packages(“horseshoe”) in the R terminal. The manual contains some examples.

Selected talks

Conferences

  • `Competing risks with time dependent clustering’ [slides] (PTA, London, September 2016)
  • `How many needles in the haystack? Adaptive inference and uncertainty quantification for the horseshoe’ [slides(JSM, Chicago, August 2016)
  • `The horseshoe and more general sparsity priors’ [slides] (EYSM, Prague, September 2015)
  • `Bayesian community detection’ [slides] (JSM, Seattle, August 2015)
  • `The horseshoe estimator: posterior concentration around nearly black vectors’ [slides] (ERCIM, Pisa, December 2014)

Seminars

  • `Conditions for posterior contraction in the sparse normal means problem’ [slides] (Bayes Club, Amsterdam, April 2016)
  • `The horseshoe prior for nearly black vectors’ [slides] (Bayes Club, Amsterdam, April 2014)

Outreach

  • `Hoe lang duurt het tot…?’ [slides]  (De Leidsche Flesch lunch seminar, Leiden, May 2016)
  • `Almost the best of three worlds’ [slides] (FMF Axioma symposium, Groningen, June 2014)
  • `Hypothese toetsen en het switch-criterium’  [slides(De Leidsche Flesch lunch seminar, Leiden, April 2014)