An asymmetric multivariate weibull distribution |
| |
Authors: | Višnja Jurić Tomasz J Kozubowski |
| |
Institution: | 1. Effectus College for Finance and Law, Zagreb, Croatia;2. Department of Mathematics &3. Statistics, University of Nevada, Reno, NV, USA |
| |
Abstract: | AbstractA class of multivariate laws as an extension of univariate Weibull distribution is presented. A well known representation of the asymmetric univariate Laplace distribution is used as the starting point. This new family of distributions exhibits some similarities to the multivariate normal distribution. Properties of this class of distributions are explored including moments, correlations, densities and simulation algorithms. The distribution is applied to model bivariate exchange rate data. The fit of the proposed model seems remarkably good. Parameters are estimated and a bootstrap study performed to assess the accuracy of the estimators. |
| |
Keywords: | Financial data modeling Gaussian mixture Laplace law Multivariate Weibull distribution Skew multivariate distribution |
|
|