BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Marginal Models for Dependent\, Clustered\, and Longitudinal Categ
 orical Data - Wicher Bergsma\, London School of Economics and Political Sc
 ience
DTSTART:20120203T160000Z
DTEND:20120203T170000Z
UID:TALK35235@talks.cam.ac.uk
CONTACT:Richard Samworth
DESCRIPTION:In the social\, behavioural\, educational\, economic\, and bio
 medical sciences\,\ndata are often collected in ways that introduce depend
 encies in the\nobservations to be compared. For example\, the same respond
 ents are\ninterviewed at several occasions\, several members of networks o
 r groups are\ninterviewed within the same survey\, or\, within families\, 
 both children and\nparents are investigated. Statistical methods that take
  the dependencies in\nthe data into account must then be used\, e.g.\, whe
 n observations at time one\nand time two are compared in longitudinal stud
 ies. At present\, researchers\nalmost automatically turn to multi-level mo
 dels or to GEE estimation to deal\nwith these dependencies. Despite the en
 ormous potential and applicability of\nthese recent developments\, they re
 quire restrictive assumptions on the\nnature of the dependencies in the da
 ta. The marginal models of this talk\nprovide another way of dealing with 
 these dependencies\, without the need for\nsuch assumptions\, and can be u
 sed to answer research questions directly at\nthe intended marginal level.
  The maximum likelihood method\, with its\nattractive statistical properti
 es\, is used for fitting the models.\n\nIn the talk I will also spend time
  on some interesting mathematical problems\nthat arise in marginal modelli
 ng. These relate to smoothness of the manifold\nof probability distributio
 ns satisfying a marginal model\, and the linear\nindependence of a set of 
 marginal parameters. This is important for the\napplicability of standard 
 asymptotic theory\, and for the calculation of the\ncorrect number of degr
 ees of freedom for a model.\n\nThis talk is based on a recent book by the 
 authors in the Springer series\nStatistics for the Social Sciences\, see w
 ww.cmm.st\n\nJoint work with Marcel Croon and Jacques Hagenaars
LOCATION:MR12\, CMS\, Wilberforce Road\, Cambridge\, CB3 0WB
END:VEVENT
END:VCALENDAR