首页> 外文期刊>SIAM Journal on Applied Mathematics >RATE OF CONVERGENCE FOR DERIVATIVE ESTIMATION OF DISCRETE-TIME MARKOV CHAINS VIA FINITE-DIFFERENCE APPROXIMATION WITH COMMON RANDOM NUMBERS
【24h】

RATE OF CONVERGENCE FOR DERIVATIVE ESTIMATION OF DISCRETE-TIME MARKOV CHAINS VIA FINITE-DIFFERENCE APPROXIMATION WITH COMMON RANDOM NUMBERS

机译:通过具有公共随机数的有限差分逼近离散时间马尔可夫链的导数估计的收敛速度

获取原文
获取原文并翻译 | 示例
           

摘要

The scheme of common random numbers (CRN) is a very popular method for variance reduction in simulation. Finite difference (FD) is a conventional technique for derivative estimation. In this paper, we examine the effectiveness of CRN for improving the rates of convergence for FD estimates over infinite horizon for homogeneous discrete-time Markov chains. For direct FD estimates, we give sufficient conditions for the effectiveness of CRN. Based on these conditions, we also suggest several simulation schemes with guaranteed effectiveness. For ratio estimates based on regenerative structures, we prove that the use of CRN is always advantageous. [References: 20]
机译:通用随机数(CRN)方案是一种非常流行的用于减少方差的方法。有限差分(FD)是用于导数估计的常规技术。在本文中,我们研究了CRN在提高均质离散时间马尔可夫链的无限范围内FD估计收敛速度方面的有效性。对于直接FD估算,我们为CRN的有效性提供了充分的条件。基于这些条件,我们还建议了几种具有保证有效性的仿真方案。对于基于再生结构的比率估算,我们证明了使用CRN始终是有利的。 [参考:20]

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