Reference : Dynamic Style Analysis with Errors in Variables
Dissertations and theses : Master of advanced studies dissertation
Business & economic sciences : Finance
http://hdl.handle.net/2268/9994
Dynamic Style Analysis with Errors in Variables
English
Bodson, Laurent mailto [Université de Liège - ULg > HEC - Ecole de gestion de l'ULg > Gestion financière >]
Jun-2007
Université de Liège
Advanced Studies Degree in Management Sciences
20
Hübner, Georges mailto
Muller, Aline mailto
Corhay, Albert mailto
[en] Style Analysis ; Kalman Filter ; Errors in Variables ; Higher Moment Estimators
[en] This paper revisits the traditional return-based style analysis (RBSA) in presence of time-varying exposures and errors in variables. We apply a selection algorithm using the Kalman filter to identify the more appropriate benchmarks and we compute their corresponding higher moment estimators (HME), i.e. the measurement error series introducing the (cross) moments of order three and four. Then, we retain the most significant HME and we add them to the selected benchmarks. Therefore, we obtain the most relevant benchmarks with none, some or all their HME as benchmarks explaining the analyzed fund return. We finally run the Kalman filter on the principal components of this set of selected benchmarks to avoid multicollinearity problems. Analysing EDHEC alternative indexes styles, we show that this technique improves the factor loadings and permits to identify more precisely the return sources of the considered fund.
Deloitte fellow
Construction d'un cadre théorique et pratique pour aborder la thématique de l'intégration d'instruments financiers complexes dans un référentiel rendement-risque à quatre moments et implications pour la gestion du risque et de portefeuille.
Researchers
http://hdl.handle.net/2268/9994

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Restricted access
WP_DEA2.pdfPublisher postprint161.3 kBRequest copy
Restricted access
WP_DEA1.docPublisher postprint38 kBRequest copy

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.