BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Wavelet-based Bayesian Estimation of Long Memory Models - an Appli
 cation to fMRI Data - Vannucci\, M (Rice University)
DTSTART:20140204T140000Z
DTEND:20140204T150000Z
UID:TALK50644@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:This talk will consider wavelet-based methods for long memory 
 estimation. Data from long memory processes have the distinctive feature t
 hat the correlation between distant observations is not negligible. Wavele
 ts\, being self-similar\, have a strong connection to long memory processe
 s and have proven to be a powerful tool for the analysis and synthesis of 
 data from such processes. Here\, in particular\, we will employ discrete w
 avelet transforms to simplify the dense\nvariance-covariance matrix of the
  error structure. We first describe a wavelet-based Bayesian procedure for
  the estimation and location of multiple change points in the long memory 
 parameter of Gaussian ARFIMA models. We then turn our attention to linear 
 regression models with long memory errors and stage a Bayesian approach to
  inference in the wavelet domain. Linear regression models with long memor
 y errors have proven useful for applications in many areas\, such as medic
 al imaging\, signal processing\, and econometrics. Recent successful appli
 cations include fMRI image data. In this talk we will consider experimenta
 l data from human cognitive tasks.\n
LOCATION:Seminar Room 2\, Newton Institute Gatehouse
END:VEVENT
END:VCALENDAR