dc.contributor |
Universitat Rovira i Virgili. Departament d'Economia |
dc.contributor |
Universitat Rovira i Virgili. Centre de Recerca en Economia Industrial i Economia Pública |
dc.contributor.author |
Aslanidis, Nektarios |
dc.contributor.author |
Martínez Ibáñez, Óscar |
dc.date.accessioned |
2012-10-29T17:07:10Z |
dc.date.available |
2012-10-29T17:07:10Z |
dc.date.created |
2012 |
dc.date.issued |
2012 |
dc.identifier.uri |
http://hdl.handle.net/2072/203167 |
dc.format.extent |
36 p. |
dc.language.iso |
eng |
dc.publisher |
Universitat Rovira i Virgili. Departament d'Economia |
dc.relation.ispartofseries |
Documents de treball del Departament d'Economia;2012-25 |
dc.rights |
info:eu-repo/semantics/openAccess |
dc.rights |
L'accés als continguts d'aquest document queda condicionat a l'acceptació de les condicions d'ús establertes per la següent llicència Creative Commons: http://creativecommons.org/licenses/by-nc-nd/3.0/es/ |
dc.source |
RECERCAT (Dipòsit de la Recerca de Catalunya) |
dc.subject.other |
Models matemàtics |
dc.title |
Modelling world investment markets using threshold conditional correlation models |
dc.type |
info:eu-repo/semantics/workingPaper |
dc.subject.udc |
33 - Economia |
dc.embargo.terms |
cap |
dc.description.abstract |
In this paper we propose a parsimonious regime-switching approach to model the
correlations between assets, the threshold conditional correlation (TCC) model. This
method allows the dynamics of the correlations to change from one state (or regime) to
another as a function of observable transition variables. Our model is similar in spirit to
Silvennoinen and Teräsvirta (2009) and Pelletier (2006) but with the appealing feature
that it does not suffer from the course of dimensionality. In particular, estimation of the
parameters of the TCC involves a simple grid search procedure. In addition, it is easy to
guarantee a positive definite correlation matrix because the TCC estimator is given by
the sample correlation matrix, which is positive definite by construction. The
methodology is illustrated by evaluating the behaviour of international equities,
govenrment bonds and major exchange rates, first separately and then jointly. We also
test and allow for different parts in the correlation matrix to be governed by different
transition variables. For this, we estimate a multi-threshold TCC specification. Further,
we evaluate the economic performance of the TCC model against a constant conditional
correlation (CCC) estimator using a Diebold-Mariano type test. We conclude that
threshold correlation modelling gives rise to a significant reduction in portfolio´s
variance. |