BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Nonparametric regression for locally stationary time series - Mich
 ael Vogt\, Department of Economics\, University of Cambridge
DTSTART:20121005T150000Z
DTEND:20121005T160000Z
UID:TALK39832@talks.cam.ac.uk
CONTACT:Richard Samworth
DESCRIPTION:We study nonparametric models allowing for locally stationary\
 nregressors and a regression function that changes smoothly over time. The
 se\nmodels are a natural extension of time series models with time-varying
 \ncoefficients. We introduce a kernel-based method to estimate the\ntime-v
 arying regression function and provide asymptotic theory for our\nestimate
 s. Moreover\, we show that the main conditions of the theory are\nsatisfie
 d for a large class of nonlinear autoregressive processes with a\ntime-var
 ying regression function. Finally\, we examine structured models\nwhere th
 e regression function splits up into time-varying additive\ncomponents. As
  will be seen\, estimation in these models does not suffer from\nthe curse
  of dimensionality.
LOCATION:MR12\, CMS\, Wilberforce Road\, Cambridge\, CB3 0WB
END:VEVENT
END:VCALENDAR