Prof. Andrew Harvey, University of Cambridge,
Date: 21 MAY 2019 at 14:30
Event location: TBA
Type: Statistics Seminars
Circular observations pose special problems for time series mod-
eling. This article shows how the score-driven approach, developed
primarily in econometrics, provides a natural solution to the difficul-
ties and leads to a coherent and uni
ed methodology for estimation,
model selection and testing. The new methods are illustrated with
hourly data on wind direction.
KEYWORDS: Circular data, diurnal e¤ects; dynamic conditional
score model; von Mises distribution; wind direction.
JEL: C22
Contact person
Prof.ssa Alessandra Luati