BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Adaptive Spectral Estimation for Nonstationary Time Series - Stoff
 er\, D (University of Pittsburgh)
DTSTART:20140117T113000Z
DTEND:20140117T121500Z
UID:TALK50031@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:We propose a method for analyzing possibly nonstationary time 
 series by adaptively dividing the time series into an unknown but finite n
 umber of segments and estimating the corresponding local spectra by smooth
 ing splines. The model is formulated in a Bayesian framework\, and the est
 imation relies on reversible jump Markov chain Monte Carlo (RJMCMC) method
 s. For a given segmentation of the time series\, the likelihood function i
 s approximated via a product of local Whittle likelihoods. The number and 
 lengths of the segments are assumed unknown and may change from one MCMC i
 teration to another.\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR