BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Monte Carlo Inference for Alpha-Stable Processes - Tatjana Lemke\,
  Fraunhofer Institute and Signal Processing and COmmunications Laboratory\
 , CUED
DTSTART:20120206T140000Z
DTEND:20120206T150000Z
UID:TALK36334@talks.cam.ac.uk
CONTACT:Rachel Fogg
DESCRIPTION:In this talk we will present a novel approach to inference in 
 previously intractable alpha-stable stochastic processes. The methods are 
 based on a Poisson sum series representation of the alpha-stable noise pro
 cess and a modified Poisson sum series representation of the alpha-stable 
 Levy process. Both representations provide a conditionally Gaussian framew
 ork\, and hence allow for the use of an auxiliary variables Monte Carlo si
 mulation scheme. To overcome the issues due to truncation of the series\, 
 residual approximations are developed. Simulations will be given of parame
 ter estimation including model parameters\, model order and alpha-stable d
 istribution parameters for discrete-time autoregressive processes driven b
 y alpha-stable innovations.
LOCATION:LR5\, Engineering\, Department of
END:VEVENT
END:VCALENDAR