BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Valid inference from non-ignorable network sampling mechanisms - S
 imon Lunagomez (University College London)
DTSTART:20161201T140000Z
DTEND:20161201T150000Z
UID:TALK69425@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:Consider a population of individuals and a&nbsp\; network that
  encodes social connections among them.&nbsp\;&nbsp\; We are interested in
  making inference on super-population estimands that are a function of bot
 h individuals&#39\; responses and of the network\, from a sample. Neither 
 the sampling frame nor the network are available. However\, the sampling m
 echanism implicitly leverages the network to recruit individuals\, thus pa
 rtially revealing social interactions among the individuals in the sample\
 , as well as their responses.&nbsp\;&nbsp\; This is a common setting that 
 arises\, for instance\, in epidemiology and healthcare\, where samples fro
 m hard-to-reach populations are collected using link-tracing mechanisms\, 
 including respondent-driven sampling. Contrary to random sampling\, the pr
 obability models of these network sampling mechanisms carry information ab
 out the estimands of interest\, such as the incidence of certain diseases 
 in the target population.&nbsp\;&nbsp\; In this work\, we study statistica
 l properties of popular network sampling mechanisms. We formulate the esti
 mation problem in terms of Rubin&#39\;s inferential framework to explicitl
 y&nbsp\;&nbsp\; account for social network structure. We then identify key
  modeling elements that lead to inferences with good frequentist propertie
 s when dealing with data collected through non-ignorable network sampling 
 mechanisms.&nbsp\;&nbsp\; We demonstrate these methods on a study of the&n
 bsp\; incidence of HIV in Brazil.  &nbsp\;  Joint work with Edoardo Airold
 i  <br><br><br>
LOCATION:Seminar Room 2\, Newton Institute
END:VEVENT
END:VCALENDAR