BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Hierarchical Evolutionary Stochastic Search with Adaptive Proposal
 s - Leonardo Bottolo\, Institute of Mathematical Sciences\, Imperial Colle
 ge London
DTSTART:20080701T133000Z
DTEND:20080701T143000Z
UID:TALK11861@talks.cam.ac.uk
CONTACT:Nikolaos Demiris
DESCRIPTION:Multivariate regression models with many responses has attract
 ed the attention of the statistical community in very recent years. A nota
 ble example is the paradigm of eQTL analysis\, where thousands of transcri
 pts are regressed versus (hundred of) thousands of markers. In this contex
 t the usual problem of multimodality of the posterior distribution\, when 
 p>>n\, is further exacerbated by the dimension of the response matrix\, us
 ually q>>n. The problem can be even more complex when transcriptomic data 
 are available for multiple tissues\, introducing a third dimension in the 
 response matrix.\n\nIn this talk we introduce a new searching algorithm ca
 lled Hierarchical Evolutionary Stochastic Search (HESS) where the response
 s are linked in a hierarchical way. To reduce the computational burden\, m
 ost of the regression parameters are integrated out. A novel sampling stra
 tegy based on Evolutionary Monte Carlo has been designed to efficiently sa
 mple from the huge parametric space. Moreover the whole set of past visite
 d models are also considered through an adaptive proposal distribution\, a
 llowing the algorithm to balance between the freedom of exploration and th
 e ability to persist on models in regions of high posterior probability.\n
 \nSimulated and real data sets are analysed to demonstrate the performance
  of the proposed algorithm when p and q are both larger than n and when mu
 ltiple tissues are considered. Collaborators on various aspects of work: S
 ylvia Richardson\, David Welsh and Enrico Petretto.
LOCATION:Large Seminar Room\, 1st Floor\, Institute of Public Health\, Uni
 versity Forvie Site\, Robinson Way\, Cambridge
END:VEVENT
END:VCALENDAR