BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//GetJobTools//Events//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:statistics-seminar-series-conditional-formulation-of-gaussian-systems
 -on-spatial-networks-with-irregularly-distributed-points-2026@getjobtools
 .com
DTSTAMP:20260924T111352Z
DTSTART:20261016T180000Z
DTEND:20261016T190000Z
SUMMARY:Statistics Seminar Series: Conditional Formulation of Gaussian Sy
 stems on Spatial Networks with Irregularly Distributed Points
DESCRIPTION:Conditional Formulation of Gaussian Systems on Spatial Networ
 ks with Irregularly Distributed Points\n\nJoin the Department of Statisti
 cs for a seminar featuring Debashis Mondal\, associate professor at the D
 epartment of Statistics\, Oregon State University. \n\nAbstract:\n\nIn th
 is talk I develop a mathematical and statistical framework for analyzing 
 spatial systems on irregularly distributed point locations. Using conditi
 onal formulations\, this framework constructs Gaussian systems on spatial
  networks with symmetric weights for both stationary and non-stationary p
 oint processes. The framework associates these systems with reversible Ma
 rkov chains and\, through Kipnis--Varadhan invariance principles\, establ
 ishes geostatistical scaling limits that recover the de Wijs process or G
 aussian free field under stationary sampling and generalize to more compl
 ex limits under inhomogeneous sampling. The framework further studies con
 vergence through Beurling--Deny decomposition\, the effects of non-unifor
 m sampling on limiting fields\, and normalization procedures based on Hun
 garian embeddings\, Sinkhorn normalization\, and density correction for r
 estoring canonical limits. It also enables scalable matrix-free computati
 on\, REML estimation\, conditional simulation\, spatial prediction\, and 
 sampling-invariant inference on large irregular spatial networks\, for wh
 ich there is currently little parallel in the geostatistical literature. 
 The talk concludes with applications to mapping soil organic carbon in Ta
 nzania\, reconstructing global seawater oxygen isotope fields\, and analy
 zing leukemia survival data from northwest England.\n\nThis seminar prese
 nts joint work with postdoctoral scholar Subhrajyoty Roy.\n\nAbout the Sp
 eaker:\n\nDebashis Mondal is an associate professor at the Department of 
 Statistics\, Oregon State University. Prior to joining Oregon State in 20
 14\, Mondal was on the statistics faculty at the University of Chicago. H
 e received his PhD. in statistics from the University of Washington and
LOCATION:Duques Hall\, 151\, United States
URL:https://calendar.gwu.edu/event/statistics-seminar-series-conditional-
 formulation-of-gaussian-systems-on-spatial-networks-with-irregularly-dist
 ributed-points
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR