Style Analysis; Kalman Filter; Errors in Variables; Higher Moment Estimators
Abstract :
[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.
Disciplines :
Finance
Author, co-author :
Bodson, Laurent ; Université de Liège - ULiège > HEC - École de gestion de l'ULiège > Gestion financière
Language :
English
Title :
Dynamic Style Analysis with Errors in Variables
Defense date :
June 2007
Number of pages :
20
Institution :
ULiège - Université de Liège
Degree :
Advanced Studies Degree in Management Sciences
Promotor :
Hübner, Georges ; Université de Liège - ULiège > HEC Recherche > HEC Recherche: Financial Management for the Future
Jury member :
Muller, Aline ; Université de Liège - ULiège > HEC Recherche > HEC Recherche: Financial Management for the Future
Corhay, Albert ; Université de Liège - ULiège > Ecole de Gestion de l'Université de Liège
Name of the research project :
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.