BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260605T054152EDT-3589Ra69R5@132.216.98.100
DTSTAMP:20260605T094152Z
DESCRIPTION:Spatio-temporal models for skewed processes\n\nIn the analysis 
 of most spatio-temporal processes in environmental studies\, observations 
 present skewed distributions. Usually\, a single transformation of the dat
 a is used to approximate normality\, and stationary Gaussian processes are
  assumed to model the transformed data. The choice of transformation is ke
 y for spatial interpolation and temporal prediction. We propose a spatio-t
 emporal model for skewed data that does not require the use of data transf
 ormation. The process is decomposed as the sum of a purely temporal struct
 ure with two independent components that are considered to be partial real
 izations from independent spatial Gaussian processes\, for each time t. Th
 e model has an asymmetry parameter that might vary with location and time\
 , and if this is equal to zero\, the usual Gaussian model results. The inf
 erence procedure is performed under the Bayesian paradigm\, and uncertaint
 y about parameters estimation is naturally accounted for. We fit our model
  to different synthetic data and to monthly average temperature observed b
 etween 2001 and 2011 at monitoring locations located in the south of Brazi
 l. Different model comparison criteria\, and analysis of the posterior dis
 tribution of some parameters\, suggest that the proposed model outperforms
  standard ones used in the literature. This is joint work with\n	Kelly C. M
 . Gonçalves (UFRJ\, Brazil) and Patríca L. Velozo (UFF\, Brazil).\n
DTSTART:20171123T183000Z
DTEND:20171123T193000Z
LOCATION:Room VCH-2820\, CA\, Université Laval
SUMMARY:Alexandra M. Schmidt\, McGill University
URL:https://www.mcgill.ca/mathstat/channels/event/alexandra-m-schmidt-mcgil
 l-university-282950
END:VEVENT
END:VCALENDAR