BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Uncertainty quantification for Geo-spatial process - Michelle Care
 y (University College Dublin)
DTSTART:20180320T133000Z
DTEND:20180320T143000Z
UID:TALK102667@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:<span>Co-author: James Ramsay		(Prof)        <br></span><br>Ge
 o spatial data are observations of a process that are collected in conjunc
 tion with reference to their geographical location. This type of data is a
 bundant in many scientific fields\, some examples include: population cens
 us\, social and demographic (health\, justice\, education)\, economic (bus
 iness surveys\, trade\, transport\, tourism\, agriculture\, etc.) and envi
 ronmental (atmospheric and oceanographic) data. They are often distributed
  over irregularly shaped spatial domains with complex boundaries and inter
 ior holes. Modelling approaches must account for the spatial dependence ov
 er these irregular domains as well as describing there temporal evolution.
  <br><br>Dynamic systems modelling has a huge potential in statistics\, as
  evidenced by the amount of activity in functional data analysis. Many see
 mingly complex forms of functional variation can be more simply represente
 d as a set of differential equations\, either ordinary or partial. <br><sp
 an><br>In this talk\, I will present a class of semi parametric regression
  models with differential regularization in the form of PDEs. This methodo
 logy will be called Data2PDE &ldquo\;Data to Partial Differential Equation
 s". Data2PDE characterizes spatial processes that evolve over complex geom
 etries in the presence of uncertain\, incomplete and often noisy observati
 ons and prior knowledge regarding the physical principles of the process c
 haracterized by a PDE.</span>
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR