BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:Probability and Statistics Seminar: Group Network Hawkes Proce
 ss
X-WR-TIMEZONE:Pacific Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20261014T134537Z
UID:tag:localist.com\,2008:EventInstance_45704123656045
DTSTART:20240226T233000Z
DTEND:20240227T002000Z
DESCRIPTION:Yongtao Guan\, Chinese University of Hong Kong\, Shenzhen\n\nAb
 stract: In this work\, we study the event occurrences of individuals inter
 acting in a network. To characterize the dynamic interactions among the in
 dividuals\, we propose a group network Hawkes process (GNHP) model whose n
 etwork structure is observed and fixed. In particular\, we introduce a lat
 ent group structure among individuals to account for the heterogeneous use
 r-specific characteristics. A maximum likelihood approach is proposed to s
 imultaneously cluster individuals in the network and estimate model parame
 ters. A fast EM algorithm is subsequently developed by utilizing the branc
 hing representation of the proposed GNHP model. Theoretical properties of 
 the resulting estimators of group memberships and model parameters are inv
 estigated under both settings when the number of latent groups G is over-s
 pecified or correctly specified. A data-driven criterion that can consiste
 ntly identify the true G under mild conditions is derived. Extensive simul
 ation studies and an application to a data set collected from Sina Weibo a
 re used to illustrate the effectiveness of the proposed methodology.
GEO:34.022409;-118.291027
LOCATION:Kaprielian Hall (KAP)\, 427
SUMMARY:Probability and Statistics Seminar: Group Network Hawkes Process
URL;VALUE=URI:https://calendar.usc.edu/event/probability_and_statistics_sem
 inar_group_network_hawkes_process
CATEGORIES:Lecture / Talk / Workshop
END:VEVENT
END:VCALENDAR