• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Advances in Computational Methods in Sciences and Engineering 2005 vol.4A; Lecture Series on Computer and Computational Sciences; vol.4A >Tuning a Portfolio Management Model on Historical Data using a Parallel Three-Stage Stochastic Programming Model
【24h】

Tuning a Portfolio Management Model on Historical Data using a Parallel Three-Stage Stochastic Programming Model

机译：使用并行三阶段随机规划模型调整历史数据的投资组合管理模型

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

A multi-stage model for allocation of financial resources to bond indices in different currencies is presented. The model is tested on historical data of interest rates and exchange rates. We consider a three-stage stochastic programming model solved by the interior point method (IPM). The application of the BQ method to the IPM allows a decomposition of the large linear system to smaller blocks allowing thus solving it in parallel. The designed parallel code is written in the Fortran/MPI programming language using the LAPACK library calls for managing the linear algebra operations on small block matrices and experiments were executed on a IMP 1350 Linux cluster. We outline the possibility of using of the program as a Grid service in the context of the Vienna Grid Environment.
机译：提出了一种将财务资源分配给不同货币的债券指数的多阶段模型。该模型在利率和汇率的历史数据上进行了测试。我们考虑通过内点法（IPM）解决的三阶段随机规划模型。 BQ方法应用于IPM可以将大型线性系统分解为较小的块，从而可以并行求解。使用LAPACK库调用以Fortran / MPI编程语言编写设计的并行代码，以管理在小块矩阵上的线性代数运算，并在IMP 1350 Linux集群上执行了实验。我们概述了在维也纳网格环境中将该程序用作网格服务的可能性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号