Purpose-The purpose of this paper is to discuss a multiscale pricing model for the French stock market by combining wavelet analysis and Fama-French three-factor model. The objective is to examine the relationship between stock returns and Fama-French risk factors at different timescales. Design/methodology/approach-Exploiting the scale separation property inherent to the maximal overlap discrete wavelet transform, the data set are decomposed into components associated with different timescales. This wavelet-based decomposition scheme allows the three Fama-French models to be tested over different investments periods. Findings-The obtained results show that the explanatory power of the Fama-French three-factor model becomes stronger as the wavelet scale increases. Besides, the relationship between the portfolio returns and the risk factors (i.e. the market, size and value factors) depends significantly upon the considered time-horizon. Practical implications-The proposed methodology offers investors the opportunity to construct dynamic portfolio management strategies by taking into account the multiscale nature of risk and return. Moreover, it gives a new insight to fund rating and fund selection issues in relation to heterogeneous investments periods. Originality/value-The paper uses wavelets as a relatively new and powerful tool for statistical analysis that allows a new understanding of pricing models. The paper will be of interest not only for academics in the field of asset pricing but also for fund managers and financial market investors.

In this paper, fractional integrating dynamics in the return and the volatility series of stock market indices are investigated. The investigation is conducted using wavelet ordinary least squares, wavelet weighted least squares and the approximate Maximum Likelihood estimator. It is shown that the long memory property in stock returns is approximately associated with emerging markets rather than developed ones while strong evidence of long range dependence is found for all volatility series. The relevance of the wavelet-based estimators, especially, the approximate Maximum Likelihood and the weighted least squares techniques is proved in terms of stability and estimation accuracy.

In this paper, a hybrid scheme for time series prediction is developed based on wavelet decomposition combined with Bayesian Least Squares Support Vector Machine regression. As a filtering step, using the Maximal Overlap Discrete Wavelet Transform, the original time series is mapped on a scale-by-scale basis yielding an outcome set of new time series with simpler temporal dynamic structures. Next, a scale-by-scale Bayesian Least Squares Support Vector Machine predictor is provided. Individual scale predictions are subsequently recombined to yield an overall forecast. The relevance of the suggested procedure is shown on the NINO3 SST anomaly index via a comparison with the existing methods for modeling and prediction.

No abstract

International audienceIn this paper, we introduce a new class of estimators of the Hurst exponent of the fractional Brownian motion (fBm) process. These estimators are based on sample expectiles of discrete variations of a sample path of the fBm process. In order to derive the statistical properties of the proposed estimators, we establish asymptotic results for sample expectiles of subordinated stationary Gaussian processes with unit variance and correlation function satisfying $\rho(i)\sim \kappa|i|^{-\alpha}$ ($\kappa\in \RR$) with $\alpha>0$. Via a simulation study, we demonstrate the relevance of the expectile-based estimation method and show that the suggested estimators are more robust to data rounding than their sample quantile-based counterparts