BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2//
METHOD:PUBLISH
X-WR-CALNAME;VALUE=TEXT:TSE
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:STANDARD
DTSTART:20251026T030000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20250330T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RDATE:20260329T020000
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.137615.field_date.0@www.tse-fr.eu
DTSTAMP:20260415T073226Z
CREATED:20250630T201001Z
DESCRIPTION:Andreas Alfons (University of Rotterdam  - Erasmus)\, “Clusterp
 ath Gaussian Graphical Modeling”\, MAD-Stat. Seminar\, Toulouse: TSE\, Sep
 tember 18\, 2025\, 11:00–12:15\, room Auditorium 3.\n\nGraphical models se
 rve as effective tools for visualizing conditional dependencies between va
 riables. However\, as the number of variables grows\, interpretation becom
 es increasingly difficult\, and estimation uncertainty increases due to th
 e large number of parameters relative to the number of observations. To ad
 dress these challenges\, we introduce the Clusterpath estimator of the Gau
 ssian Graphical Model (CGGM) that encourages variable clustering in the gr
 aphical model in a data-driven way. Through the use of an aggregation pena
 lty\, we group variables together\, which in turn results in a block-struc
 tured precision matrix whose block structure remains preserved in the cova
 riance matrix. The CGGM estimator is formulated as the solution to a conve
 x optimization problem\, making it easy to incorporate other popular penal
 ization schemes which we illustrate through the combination of an aggregat
 ion and sparsity penalty. We present a computationally efficient implement
 ation of the CGGM estimator by using a cyclic block coordinate descent alg
 orithm. In simulations\, we show that CGGM not only matches\, but oftentim
 es outperforms other state-of-the-art methods for variable clustering in g
 raphical models. We also demonstrate CGGM's practical advantages and versa
 tility on a diverse collection of empirical applications.
DTSTART;TZID=Europe/Paris:20250918T120000
DTEND;TZID=Europe/Paris:20250918T131500
LAST-MODIFIED:20260113T095129Z
LOCATION:Toulouse: TSE\, September 18\, 2025\, 11:00–12:15\, room Auditoriu
 m 3
SUMMARY:MAD-Stat. Seminar
URL;TYPE=URI:https://www.tse-fr.eu/seminars/2025-clusterpath-gaussian-graph
 ical-modeling
END:VEVENT
END:VCALENDAR