David Rossell

Biosketch

I obtained my PhD from Rice University (Houston, USA) under Peter Mueller, working on Bayesian methodology for sequential design of multi-arm and massive multiple hypothesis testing experiments, and on the analysis of differential gene expression.

I did a post-doc under Valen Johnson and Veera Badalandayuthapani at MD Anderson Cancer Center (Houston, USA) working on Bayesian model selection, including the proposal of non-local priors.

I moved on to the Institute for Research in Biomedicine of Barcelona (Spain) to create and lead a Biostatistics & Bioinformatics Unit for about 6 years. There I developed statistical methodology for high-dimensional problems and for a variety of bioinformatics applications requiring a delicate balance between theoretical soundness and computational scalability, also I was involved in a large number of inter-disciplinary collaborations in biology and chemistry featured in the top journals of those fields.  

In 2013 I moved to the Statistics Dept. at the University of Warwick (Coventry, UK) as assistant and later associate professor, where I deepened my studies of some methodological aspects of Bayesian model selection and averaging, as well as continuing inter-disciplinary work.

In 2016 I moved to Universitat Pompeu Fabra as a Ramón y Cajal Fellow in Mathematics, where I continue to work on methodological issues such as theory and computation for high-dimensional inference, and engaging in a number of exciting problems in the social sciences and economics such as understanding massive trading networks, real state prices, or the use of social media by extremist political parties.

Education

• 2003 - 2006. PhD in Biostatistics (GPA 4.08) Rice University (Houston, USA)
• 1999 - 2001. Llicenciatura in Statistics. Universitat Polite?cnica de Catalunya (Barcelona, Spain)
• 1996 - 1999. Diplomatura in Statistics. Universitat Polite?cnica de Catalunya (Barcelona, Spain)

 

Appointments

2016 -present
Universitat Pompeu Fabra, Dept. of Economics & Business (Barcelona, Spain). 
Visiting professor and Co-director of GSE Barcelona Data Science Center

2013-2016
University of Warwick, Dept. of Statistics (Coventry, UK) 
Assistant professor (2013-2016) and Associate professor (2016)

2008-2013
Institute for Research in Biomedicine (Barcelona, Spain) 
Head of the Biostatistics & Bioinformatics Unit

2007
MD Anderson Cancer Center, Biostatistics Dept. (Houston, USA) 
Post-doctoral fellow

1999 - 2003 
Biome?trica Me?dica i Social (Barcelona, Spain) 
Data analyst (1999-2001), Head of Statistics Department (2001-2003)

Research lines

Bayesian statistics, high-dimensional inference, model selection, experimental design, applications in biomedicine/economics/social sciences.

Selected publications

Rossell D, Rubio FJ. Tractable Bayesian variable selection: beyond normality (2017). Journal of the American Statistical Association (DOI 10.1080/01621459.2017.1371025).

Papaspiliopoulos O., Rossell D. Bayesian block-diagonal variable selection and model averaging. (2017). Biometrika, 104, 343-359.

Rossell D., Telesca D. Non-local priors for high-dimensional estimation (2017). Journal of the American Statistical Association, 112, 254-265.

Johnson V.E., Rossell D. (2012) Bayesian model selection in High-dimensional settings. Journal of the American Statistical Association, 107, 498, 649-660

Johnson V.E., Rossell D. (2010) Non-Local Prior Densities for Default Bayesian Hypothesis Tests. Journal of the Royal Statistical Society B, 72, 143-170