Plenary Lecture
Wavelet Based Risk Measures
Professor Rossitsa Yalamova
University of Lethbridge, Canada
E-mail: rossitsa.yalamova@uleth.ca
Abstract: This research tests two wavelet based risk measures. The advantages of wavelet methodologies are multiple scale analysis and no need for a priori assumptions about the data distribution. While the first of the methodologies measures wavelet-based realized volatility at multiple time scale, the second additionally accounts for the higher order cumulants. A comparison of the two methods quantifies the horizon trade-offs for portfolio optimization. Wavelet-based estimators have been used very successfully for estimating scaling behavior. These estimators are blind to superimposed polynomial trends and are also very robust when the shape of the underlying distribution is changed (Abry et al.1998). The Wavelet Transform Modulus Maxima (WTMM) method allows us to build an estimator that is based on the local maxima of the continuous wavelet transform. This method has proven very efficient to compute the singularity spectrum of multifractal signals. And as shown by Audit et al. (2002) a WTMM estimator is a very good candidate for analyzing real data without any prior assumptions regarding the distribution type.
Brief Biography of the Speaker: CEEL program in Adaptive Economic Dynamics, University of Trento 2012
MSRI Berkeley Workshop Percolation & Interactive Systems 2012
University of Brunei Darussalam/IBM Global Sustainability Summer School 2011
Santa Fe Institute & Institute of Theoretical Physics Beijing
Complex Systems Summer School 2007
Kent State University PhD Finance 2003
University of Pittsburgh, Katz Graduate School of Business MBA 1995
University Of Sofia Bulgaria; ABD microbiology 1991
Saint Petersburg State Medical Academy, Russia MD 1985
More Details