BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Special ENAC Research Day & EESS talk on "Multivariate Time Series
  Models for Environmental Applications based on Latent Stochastic Processe
 s"
DTSTART:20170523T124500
DTEND:20170523T134500
DTSTAMP:20260510T045224Z
UID:da698897652856b88bbcc2e255198b057a372b0143f03ce34403f20d
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr Stéphane Guerrier\, Assistant Professor\, Department of St
 atistics\, University of Illinois at Urbana-Champaign (USA)\n\nShort biogr
 aphy:\nStéphane Guerrier received a M.Sc. degree in Environmental Engine
 ering from the École Polytechnique Fédérale de Lausanne (EPFL)\, Swi
 tzerland\, in 2008\, and a Ph.D. degree in Statistics from the University 
 of Geneva\, Switzerland\, in 2013. He is currently Assistant Professor wit
 h the Department of Statistics at the University of Illinois at Urbana–C
 hampaign\, IL\, USA. His current research interests include time series an
 alysis\, spatial statistics\, simulation-based estimation methods and robu
 st statistics.\nAbstract:\nA growing area of interest in ecological resear
 ch lies in the effects of changes in climate on biological communities and
  the effects of changes in biological communities on weather and climate. 
 Climate conditions often are measured by ecologists as time series either 
 with meteorological measurements recorded at weather stations or with envi
 ronmental sensors at fixed time intervals. Mixed models are a good candida
 te to deal with these types of measurements allowing to obtain a better in
 ference on the true parameters of interest for the researcher. However\, t
 he level of complexity which can be handled for the dependence structure i
 n the residuals is limited in practice. Indeed\, the examples in ecologica
 l research show how there can be a series of factors contributing to the d
 ependence structure in residuals over time and the observations can be the
  result of different combinations of stochastic processes. Given these con
 siderations\, there is a need to model multivariate time series with a com
 plex structure delivered by the fact that the individual time series can s
 hare some models and factors with some time series but not with others (or
  eventually not share any model at all). As a solution to this problem we 
 present a method which addresses these issues based on the concept of late
 nt processes which can be extended to a multivariate setting where the dep
 endence between observations between and within factors can be explained b
 y "shared'' models between time series. An approach based on this concept 
 appears reasonable and indeed more appropriate considering the actual cond
 itions under which experimental data is collected.  In this work we use t
 he idea of the Generalized Method of Wavelet Moments (GMWM) to estimate th
 ese multivariate time series with complex dependence structures. This is d
 one in a straightforward and computationally efficient manner allowing to 
 deliver an easy-to-implement method for the estimation of these random eff
 ects. When applied to some vapor pressure deficit data\, the method provid
 es results which are consistent with ecological interpretations and gives 
 additional insight into the dependence structures among explanatory factor
 s. This is a joint work with Allison Gardner\, Roberto Molinari and James 
 Balamuta.
LOCATION:SG 0213 https://plan.epfl.ch/theme/generalite_thm_v2?dim_floor=0&
 lang=en&dim_lang=en&tree_groups=centres_nevralgiques%2Cacces%2Cmobilite_re
 duite%2Censeignement%2Ccommerces_et_services%2Cvehicules&tree_group_l
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR