Nome Egidi Leonardo
30
XXX
Dottore di ricerca
Department of Statistical Sciences University of Padua
si
2014-2017 (Currently postdoctoral research fellow among DEAMS, University of Trieste)
legidi@units.it
2014 - 2017
Nicola Torelli (Univ. Trieste)
Francesco Pauli (Univ. Trieste)
March 2018 PhD in Statistical Sciences among the Department of Statistical Sciences, University of Padova
,September 2011- March 2014: Master degree in Actuarial and Statistical Sciences, Department of Economical, Mathematical and Statistical Sciences (DEAMS), University of Trieste
,October 2008- July 2011: Bachelor degree in Statistics, Economics and Finance at the Faculty of Economy, University of Trieste
Relabelling in Label switching problem
Bayesian hierarchical models for foorball data
Data-dependent priors
Clustering
Algorithms for sparse matrixes
January 2016 - July 2016 : Visiting Scholar among the Department of Statistics, Columbia University, New York. Supervisor: prof Andrew Gelman
April 2012- January 2013: Tutor for Bachelor's students at the Department of Economical, Mathematical and Statistical Sciences of the University of Trieste, Italy. (Financial Mathematics, Descriptive Statistics)
September 2017: Introduction to Bayesian Data Analysis with Stan, 21 hours, Frontal lectures+labs, 21 hours, Joint Research Center (JRC), Sevilla. Instructor: Jonah Gabry (Department of Statistics, Columbia University);
March-June 2018: Statistical Methods for Data Sciences (12 hours), Degree in Data Science and Scientific Computing;
December 2017: Bayesian inference: methods and applications. (Corso ARPA Friuli Venezia-Giulia)
M. Sc. Thesis Title: A comparison between Bayesian hierarchical models for the meta-analysis of data from pre-electoral polls. Supervisor: Nicola Torelli. Co-supervisor: Francesco Pauli.
Score: 110/110 cum laude
,B. Sc. Thesis Title: Consideration about the arbitrage principle. Supervisor: Marco Zecchin.
Score: 106/110
Developments in Bayesian Hierarchical Models and Prior Specification with Application to Analysis of Soccer Data. Supervisor: Nicola Torelli. Co-supervisor: Francesco Pauli.
Conference Proceedings
Egidi L., Pappada R., Pauli F. and Torelli N. - Relabelling in Bayesian Mixture Models by Pivotal Units - Una procedura di relabelling in modelli mistura Bayesiani basata su unita pivotali. Conference Paper, SIS2016 Proceedings. ISBN 9788861970618. [48th Scientic Meeting of the Italian Statistical Society, Salerno, 8-10 June 2016].
Egidi L., Gabry J.S. (2017). Bayesian hierarchical models for predicting individual performance in football (soccer). Conference Paper, {MathSport International 2017 Conference}. ISBN 978-88-6938-058-7. [Padova, 26-28 June 2017].
Journal Papers
Egidi L., Pappada R., Pauli F. and Torelli N. - Relabelling in Bayesian Mixture Models by Pivotal Units. Statistics and Computing, 28(4), 957-969, 2017, DOI 10.1007/s11222-017-9774-2.
Egidi L., Pappada R., Pauli F. and Torelli N. - Maxima units search (MUS) algorithm: methodology and applications. Studies in Theoretical and Applied Statistics, Springer Book, 2018.
Egidi L., Pauli F. and Torelli N. - Combining historical data and bookmakers' odds in modelling football scores. Submitted to Statistical Modelling, 2018
Egidi L., Pauli F. and Torelli N. - Mixture data-dependent priors. Submitted to Bayesian Analysis, 2018.
Egidi L., Gabry J.S. (2017). Bayesian hierarchical models for predicting individual performance in football (soccer). Submitted to Journal of Quantitative Analysisi in Sports
,
Talks and poster presentation
Isba 2016 (Santa Margherita di Pula)
MBC2 (Catania)
SISBayes 2017 (Roma)
BISP 10/p>
MathSport International 2017 ConferenceAUEB Sports Analytics Workshop 2017
,
Postdoctoral Research Fellow In Statistics, University of Trieste - DEAMS
