BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Isaac Newton Institute Seminar Series
SUMMARY:Wavelet-based Bayesian Estimation of Long Memory M
 odels - an Application to fMRI Data - Vannucci\, M
  (Rice University)
DTSTART;TZID=Europe/London:20140204T140000
DTEND;TZID=Europe/London:20140204T150000
UID:TALK50644AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/50644
DESCRIPTION:This talk will consider wavelet-based methods for 
 long memory estimation. Data from long memory proc
 esses have the distinctive feature that the correl
 ation between distant observations is not negligib
 le. Wavelets\, being self-similar\, have a strong 
 connection to long memory processes and have prove
 n to be a powerful tool for the analysis and synth
 esis of data from such processes. Here\, in partic
 ular\, we will employ discrete wavelet transforms 
 to simplify the dense\nvariance-covariance matrix 
 of the error structure. We first describe a wavele
 t-based Bayesian procedure for the estimation and 
 location of multiple change points in the long mem
 ory parameter of Gaussian ARFIMA models. We then t
 urn our attention to linear regression models with
  long memory errors and stage a Bayesian approach 
 to inference in the wavelet domain. Linear regress
 ion models with long memory errors have proven use
 ful for applications in many areas\, such as medic
 al imaging\, signal processing\, and econometrics.
  Recent successful applications include fMRI image
  data. In this talk we will consider experimental 
 data from human cognitive tasks.\n
LOCATION:Seminar Room 2\, Newton Institute Gatehouse
CONTACT:Mustapha Amrani
END:VEVENT
END:VCALENDAR