BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Isaac Newton Institute Seminar Series
SUMMARY:Mixed Effects Model on Functional Manifolds / Samp
ling Directed Networks - Jingjing Zou (University
of Cambridge)
DTSTART;TZID=Europe/London:20180322T090000
DTEND;TZID=Europe/London:20180322T100000
UID:TALK102787AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/102787
DESCRIPTION:I would like to talk about two projects. Co-author
s of Mixed Effects Model on Functional Manifolds:
John Aston (University of Cambridge)\, Lexin Li (U
C Berkeley) We propose a generalized mixed effects
model to study effects of subject-specific covari
ates on geometric and functional features of the s
ubjects'\; surfaces. Here the covariates includ
e both time-invariant covariates which affect both
the geometric and functional features\, and time-
varying covariates which result in longitudinal ch
anges in the functional textures. In addition\, we
extend the usual mixed effects model to model the
covariance between a subject'\;s geometric def
ormation and functional textures on the surface.
Co-authors of Sampling Directed Networks: Richard
Davis (Columbia University)\, Gennady Samorodnits
ky (Cornell University)\, Zhi-Li Zhang (University
of Minnesota). We propose a sampling procedure f
or the nodes in a network with the goal of estimat
ing uncommon population features of the entire net
work. Such features might include tail behavior o
f the in-degree and out-degree distributions and a
s well as their joint distribution. Our procedure
is based on selecting random initial nodes and th
en following the path of linked nodes in a structu
red fashion. In this procedure\, targeted nodes w
ith desired features\, such as large in-degree\, w
ill have a larger probability of being retained.
In order to construct nearly unbiased estimates of
the quantities of interest\, weights associated w
ith the sampled nodes must be calculated. We will
illustrate this procedure and compare it with a s
ampling scheme based on multiple random walks on s
everal data sets including webpage network data an
d Google+ social network data.
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR