Regime Switching Models for Directional and Linear Observations
Preprint
Repository URI
Repository DOI
Change log
Authors
Harvey, A.
Palumbo, D.
Abstract
The score-driven approach to time series modeling provides a solution to the problem of modeling circular data and it can also be used to model switching regimes with intra-regime dynamics. Furthermore it enables a dynamic model to be fitted to a linear and a circular variable when the joint distribution is a cylinder. The viability of the new method is illustrated by estimating a model with dynamic switching and dynamic location and/or scale in each regime to hourly data on wind direction and speed in Galicia, north-west Spain.
Description
Keywords
Circular data, conditional score, cylinder, hidden Markov model, von Mises distribution, wind
Is Part Of
Publisher
Faculty of Economics, University of Cambridge