A metaheuristic approach for correlated random vector generation
The generation of correlated random variables is relevant in the stochastic simulation of financial and manufacturing systems, among many other applications. The generally accepted techniques to generate correlated multivariate observations rely on the mathematical attributes of the probability dens...
- Autores:
-
Hurtado Medina, Edgard Mauricio
Guaje Acosta, Oscar Orlando
- Tipo de recurso:
- Fecha de publicación:
- 2019
- Institución:
- Universidad de los Andes
- Repositorio:
- Séneca: repositorio Uniandes
- Idioma:
- eng
- OAI Identifier:
- oai:repositorio.uniandes.edu.co:1992/44349
- Acceso en línea:
- http://hdl.handle.net/1992/44349
- Palabra clave:
- Correlación (Estadística) - Investigaciones
Generador de números aleatorios - Métodos de simulación - Investigaciones
Optimización matemática - Investigaciones
Ingeniería
- Rights
- openAccess
- License
- https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf
| Summary: | The generation of correlated random variables is relevant in the stochastic simulation of financial and manufacturing systems, among many other applications. The generally accepted techniques to generate correlated multivariate observations rely on the mathematical attributes of the probability density functions of the random variables. In this paper, we propose a new approach based on Iterated Local Search (ILS) that induces a desired correlation structure to multivariate random data independent of the probability density function of the input variables. The proposed methodology improves the quality of the results found by the Iman and Conover method - currently used in commercial simulators such as Crystal Ball - at a low computational cost. |
|---|