Publication:
Modified Whittle estimation of multilateral models on a lattice

Simple item page
dc.affiliation.dptoUC3M. Departamento de Economía de la Empresaes
dc.contributor.authorRobinson, P.M.
dc.contributor.authorVidal-Sanz, Jose M.
dc.date.accessioned2010-04-12T14:44:00Z
dc.date.available2010-04-12T14:44:00Z
dc.date.issued2006
dc.description.abstractIn the estimation of parametric models for stationary spatial or spatio-temporal data on a d-dimensional lattice, for d≥2, the achievement of asymptotic efficiency under Gaussianity, and asymptotic normality more generally, with standard convergence rate, faces two obstacles. One is the “edge effect”, which worsens with increasing d. The other is the possible difficulty of computing a continuous-frequency form of Whittle estimate or a time domain Gaussian maximum likelihood estimate, due mainly to the Jacobian term. This is especially a problem in “multilateral” models, which are naturally expressed in terms of lagged values in both directions for one or more of the d dimensions. An extension of the discrete-frequency Whittle estimate from the time series literature deals conveniently with the computational problem, but when subjected to a standard device for avoiding the edge effect has disastrous asymptotic performance, along with finite sample numerical drawbacks, the objective function lacking a minimum-distance interpretation and losing any global convexity properties. We overcome these problems by first optimizing a standard, guaranteed non-negative, discrete-frequency, Whittle function, without edge-effect correction, providing an estimate with a slow convergence rate, then improving this by a sequence of computationally convenient approximate Newton iterations using a modified, almost-unbiased periodogram, the desired asymptotic properties being achieved after finitely many steps. The asymptotic regime allows increase in both directions of all d dimensions, with the central limit theorem established after re-ordering as a triangular array. However our work offers something new for “unilateral” models also. When the data are non-Gaussian, asymptotic variances of all parameter estimates may be affected, and we propose consistent, non-negative definite estimates of the asymptotic variance matrix.
dc.description.statusPublicado
dc.format.mimetypeapplication/pdf
dc.format.mimetypetext/plain
dc.identifier.bibliographicCitationJournal of Multivariate Analysis, 2006, 97, 5, p. 1090–1120
dc.identifier.doi10.1016/j.jmva.2005.05.013
dc.identifier.issn0047-259X
dc.identifier.publicationfirstpage1090
dc.identifier.publicationissue5
dc.identifier.publicationlastpage1120
dc.identifier.publicationtitleJournal of Multivariate Analysis
dc.identifier.publicationvolume97
dc.identifier.urihttps://hdl.handle.net/10016/7249
dc.language.isoeng
dc.publisherElsevier
dc.relation.publisherversionhttp://dx.doi.org/10.1016/j.jmva.2005.05.013
dc.rights©Elsevier
dc.rights.accessRightsopen access
dc.subject.ecienciaEmpresa
dc.subject.otherSpatial data
dc.subject.otherMultilateral modelling
dc.subject.otherWhittle estimation
dc.subject.otherEdge effect
dc.subject.otherConsistent variance estimation
dc.titleModified Whittle estimation of multilateral models on a lattice
dc.typeresearch article*
dc.type.reviewPeerReviewed
dspace.entity.typePublication
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
whittle_JMA_2006_ps.pdf
Size:
445.55 KB
Format:
Adobe Portable Document Format
Description:
Download
Collections
DEE - Artículos de Revistas
Economists Online