Show simple item record

dc.contributor.authorHarvey, A.
dc.contributor.authorHurn, S.
dc.contributor.authorThiele, S.
dc.date.accessioned2019-09-17T09:00:05Z
dc.date.available2019-09-17T09:00:05Z
dc.date.issued2019-08-12
dc.identifier.otherCWPE1971
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/296873
dc.description.abstractCircular observations pose special problems for time series modeling. This article shows how the score-driven approach, developed primarily in econometrics, provides a natural solution to the difficulties and leads to a coherent and unified methodology for estimation, model selection and testing. The new methods are illustrated with hourly data on wind direction.
dc.publisherFaculty of Economics
dc.relation.ispartofseriesCambridge Working Papers in Economics
dc.rightsAll Rights Reserved
dc.rights.urihttp://www.rioxx.net/licenses/all-rights-reserved/
dc.subjectAutoregression
dc.subjectcircular data
dc.subjectdynamic conditional score model
dc.subjectvon Mises distribution
dc.subjectwind direction
dc.titleModeling directional (circular) time series
dc.typeWorking Paper
dc.identifier.doi10.17863/CAM.43915


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record