Authors: Anyssa Trimech; Saloua Benammou
Addresses: Computational Mathematics Laboratory, Faculty of Law, Economic, and Political Sciences, Sousse University, Rue 2 mars, Kheniss, Monastir 5011, Sousse, Tunisia. ' Computational Mathematics Laboratory, Faculty of Law, Economic, and Political Sciences, Sousse University, Rue 2 mars, Kheniss, Monastir 5011, Sousse, Tunisia
Abstract: The purpose of this paper is to consider the utilities of using value at risk (VaR) as an explanatory factor of stock returns in addition to the Fama-French risk factors over different investment periods. In order to describe the relationships between factors and stock returns and to examine the explanatory power of the four factor model at different timescales, we exploit the properties of the multi-resolution analysis (MRA) based on maximal overlap discrete wavelet transform (MODWT). The four factor model proposed by Turan G. Bali and Nusret Cakici illustrates well the cross-sectional returns while the timescale increases. The portfolio returns are more sensitive to the market risk and size risk. The portfolio risk effect, measured by the VaR, is handed-over in question because its weakness and the addressed criticism following the current financial crisis.
Keywords: multifactor asset pricing models; value at risk; VaR; three factor models; Eugene Fama; Kenneth French; stock returns; asset pricing; portfolio management; risk factors; MRA; multiresolution analysis; explanatory factor analysis; France; investment periods; explanatory powers; four factor models; MODWT; maximal overlap discrete wavelet transform; Turan Bali; Nusret Cakici; cross-sectional returns; timescales; portfolio returns; market risk; size risk; portfolio risk effects; financial crises; decision sciences; risk management.
International Journal of Decision Sciences, Risk and Management, 2012 Vol.4 No.1/2, pp.58 - 76
Received: 01 Dec 2010
Accepted: 12 Jul 2011
Published online: 02 May 2012 *