Lindley Prize Awarded to Wade, Mongelluzzo and Petrone
The International Society for Bayesian Analysis (ISBA), the most important scientific society for Bayesian statistics, has awarded the Lindley Prize 2010 to Sara Wade, Silvia Mongelluzzo and Sonia Petrone (Department of Decision Sciences), for their paper An Enriched Conjugate Prior for Bayesian Nonparametric Inference (Bayesian Analysis, Volume 6, Number 3, September 2011, Pages 359-500). This important achievement is particularly welcome since Wade and Mongelluzzo are graduate students of the PhD in Statistics at Bocconi. The prize awarding took place at the ISBA 2012 World Meeting last July, in Kyoto, Japan.
ISBA promotes the development and application of Bayesian analysis useful in the solution of theoretical and applied problems in science, industry and government. The Lindley Prize is awarded for innovative research in Bayesian Statistics that is presented at an ISBA World Meeting. Award winning papers have to present research in Bayesian statistics that is judged important, timely and notably original.
The paper by Wade, Mongelluzzo and Petrone proposes an extension of the Dirichlet process, a widely used process in Bayesian nonparametric statistics. "Our contribution", Petrone explains, "enriches the Dirichlet process and makes it more flexible, succeeding in maintaining it simple. Simplicity and flexibility are hard to find together".
As Petrone says, "Bayesian nonparametrics (BNP) is a research field studied in a pioneering way at Bocconi since the late 70's, with professors Michele Cifarelli and Eugenio Regazzini. In the last decade, BNP has experienced an impressive development, with an active and fruitful interaction with other academic communities: economics, genetics, biostatistics, social networks and especially machine learning". Thanks to this award-winning contribution all these communities have a novel, more flexible tool for applications of BNP in complex models.