BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20251110T020118EST-8257k5861M@132.216.98.100
DTSTAMP:20251110T070118Z
DESCRIPTION:Title: How close and how much? Linking health outcomes to spati
 al distributions of built environment features.\n\nAbstract: Veronica Berr
 ocal is an Associate Professor in the Department of Statistics at UC Irvin
 e (UCI). She joined UCI in Fall 2019\, after having spent 9 years as facul
 ty in the Department of Biostatistics at University of Michigan. Veronica 
 earned her Ph.D. in Statistics at University of Washington in 2007\, and w
 as a postdoctoral fellow in the Research Triangle Park area for 3 years\, 
 performing research first for a year at the US EPA as a National Research 
 Council postdoctoral associate\, and then for two years at Duke University
 /SAMSI. Veronica's research interests are in spatial/spatio-temporal stati
 stics\, statistical methods for environmental epidemiology and environment
 al exposure assessment\, with a particular focus on air pollution\, climat
 e\, weather\, and more recently the built environment\, in particular as t
 hey relate to human health.\n\n\nBuilt environment features (BEFs) refer t
 o aspects of the human constructed environment\, which may in turn support
  or restrict health related behaviors and thus impact health. In this talk
  we will present two approaches to understand whether the spatial distribu
 tion and quantity of fast food restaurants (FFRs) influence BMI and the ri
 sk of obesity in schoolchildren. The first method is focused on determinin
 g the “radius of influence” of BEFs on children’s BMI values and extends t
 he class of Distributed Lag Models to the spatial context. The second meth
 od is a two-stage Bayesian hierarchical modeling framework that examines h
 ow the spatial pattern and quantity of FFRs affect the risk of obesity. Th
 e first stage of the hierarchical model uses the position of FFRs relative
  to that of some reference locations - in our case\, schools - and models 
 them as realizations of 1-dimensional Inhomogenous Poisson processes (IPP)
 . With the goal of identifying representative spatial patterns of exposure
  to FFRs\, we adopt a Bayesian non-parametric viewpoint and provide the in
 tensity functions of the IPPs with a Nested Dirichlet Process prior. The s
 econd stage model relates exposure patterns to obesity\, offering two diff
 erent approaches to accommodate uncertainty in the exposure patterns estim
 ated in the first stage.\n	\n	Our analysis on the influence of patterns of F
 FR occurrence on obesity among Californian schoolchildren has indicated th
 at\, in 2010\, among schools that are consistently assigned to a cluster\,
  the odds of obesity amongst 9th graders who attend schools with most dist
 ant FFR occurrences in a 1-mile radius are lower than for other schoolchil
 dren.\n\nFor Zoom meeting : Please contact: admincoord.eboh [at] mcgill.ca
 \n\n \n
DTSTART:20220223T203000Z
DTEND:20220223T213000Z
SUMMARY:Veronica Berrocal (University of California\, Irvine)
URL:/mathstat/channels/event/veronica-berrocal-univers
 ity-california-irvine-337859
END:VEVENT
END:VCALENDAR