BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
TZID:Europe/Paris
X-WR-TIMEZONE:Europe/Paris
BEGIN:VEVENT
UID:6087@i2m.univ-amu.fr
DTSTART;TZID=Europe/Paris:20220408T143000
DTEND;TZID=Europe/Paris:20220408T153000
DTSTAMP:20241120T200910Z
URL:https://www.i2m.univ-amu.fr/evenements/matched-mismatched-and-semipara
 metric-inference-in-elliptical-distributions/
SUMMARY:Stefano FORTUNATI (L2S\, Centrale Supelec\, Gif-sur-Yvette): Matche
 d\, mismatched and semiparametric inference in elliptical distributions
DESCRIPTION:Stefano FORTUNATI: Any scientific experiment\, which aims to ga
 in some knowledge about a real-word phenomenon\, starts with the data coll
 ection. In statistics and signal processing\, all the available knowledge 
 about a physical phenomenon of interest is summarized in the probability d
 ensity function (pdf) of the collected observations. To this end\, we have
  to define a model\, that is a family of pdfs that are able to statistical
 ly characterize the observations. The most used class of models are the pa
 rametric ones which however require the perfect match between the actual d
 ata distribution and the assumed model itself. Nevertheless\, in practice\
 , a certain amount of mismatch is often inevitable. Therefore\, being awar
 e about the possible performance loss that the derived estimator could und
 ergone under model misspecification is of crucial importance. Even more im
 portant would be the possibility to overcome this misspecification problem
 . This can be achieved by adopting the more general semiparametric charact
 erization of the statistical behavior of the collected data.\nIn this semi
 nar we use theset of elliptical distribution as “fil rouge” to analyse
 s the three above-mentionedaspects. In particular\, we start with an intro
 duction about classical parametric inference in the ellipticalmodel\, with
  particular attention to the Maximum Likelihood (ML) estimator ofthe data 
 covariance matrix. Then\, the quantification of the performance lossdue to
  a wrong Gaussianity assumption on the data distribution will beaddressed.
  Finally\, a new class of distributionally robust and semiparametric effic
 ient R-estimatorsof the data covariance will be presented and their perfor
 mance compared withthe one of the more “classical” M-estimators.\nKeyw
 ords: Semiparametric inference\, data mismatch\, M-estimators\, R-estimato
 rs.\nShortbio: Stefano FORTUNATI received the graduate degree in telecommu
 nication engineering and the Ph.D. degree\, both from the University of Pi
 sa\, Italy\, in 2008 and 2012\, respectively. In 2012\, he joined the Depa
 rtment of Ingegneria del l'Informazione\, University of Pisa\, where he wa
 s a researcher with a postdoc position until September 2019. Since October
  2019\, he is an associate researcher at Université Paris-Saclay\, CNRS\,
  CentraleSupélec\, Laboratoire des Signaux et Systems (L2S)\, 91190\, Gif
 -sur-Yvette\, France. From Sept. 2020 he is a permanent lecturer (enseigna
 nt-chercheur) at IPSA in the Parisian campus of Ivry-sur-Seine. From Septe
 mber 2012 to November 2012 and from September 2013 to November 2013\, he w
 as a Visiting Researcher with the CMRE NATO Research Center\, La Spezia\, 
 Italy. From May 2017 to April 2018\, he spent a period of one year as a Vi
 siting Researcher with the Signal Processing Group\, Technische Universit
 ät Darmstadt. His professional expertise encompasses different areas of t
 he statistical signal processing and applied statistics\, with particular 
 focus on point estimation and hypothesis testing\, performance bounds\, mi
 sspecification theory\, robust and semiparametric statistics and statistic
 al learning theory.\n 
ATTACH;FMTTYPE=image/jpeg:https://www.i2m.univ-amu.fr/wp-content/uploads/2
 022/04/Stefano_Fortunati.png
CATEGORIES:Séminaire,Signal et Apprentissage
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:DAYLIGHT
DTSTART:20220327T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
END:VCALENDAR