Multivariate Generalized Distributions
Sale page: amazon.com
About this book
A new class of multivariate distributions is proposed which extends the univariate class of generalized exponential distributions. This family of distributions is used to develop new classes of multivariate GARCH models and applied to a single index conditional capital asset pricing model over the period 1988 to 1995. Special attention is given to extending the stan0 dard multivariate conditional distributions of normality to higher order moments which can admit both skewness and kurtosis.
This paper builds on the ARCH approach for modeling distributions with time-varying conditional variance by using the generalized Student t distribution. The distribution offers flexibility in modeling both leptokurtosis and asymmetry
Author: Thomas McCurdy
Tom McCurdy holds the Bonham Chair in International Finance at the Rotman School of Management, University of Toronto. Tom’s recent research focuses on developing and applying quantitative methods for forecasting the dynamics of asset return distributions. He has a particular interest in sources of volatility and skewness, as well as the measurement and management of risk that results from the changing shape of return distributions. He has also co-authored over 45 learning-by-doing simulation cases designed to practice deriving robust strategies for decisions associated with topics in market microstructure, valuation, derivatives, commodities, portfolio and risk management.