Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
6895366 | European Journal of Operational Research | 2018 | 39 Pages |
Abstract
New multivariate risk measures are introduced, suitable for optimal management of multidimensional assets. Risk is measured along lines through a given reference point in a multidimensional Euclidean space, and then maximum (minimum in financial planning) or mixture is taken with respect to lines lying in cones. We use VaR and CVaR as univariate risk measures but the construction allows for the use any of them. In some case numéraire is used to value the assets. Some of the new measures enjoy the coherence property for sums and also for composition, where assets are put together to form higher dimensional vectors. Numerical calculations of them are tractable as shown for certain multivariate distributions. Applications are presented for the agricultural industry using USDA database, as well as a financial portfolio problem using recent US stock market data.
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Science (General)
Authors
András Prékopa, Jinwook Lee,