重要性采样研究进展

A review of Importance Sampling research

重要性采样算法作为蒙特卡罗仿真中一种有效的快速仿真技术,主要作用在于降低给定仿真估计器的方差,从而减少仿真样本数,缩短仿真时间。本文对重要性采样相关研究进展情况进行了分析归纳,简要阐述了研究重要性采样的目的和基本原理,对重要性密度函数的选取方法进行了分析总结,给出了重要性采样的一些典型应用成果,最后提出了重要性采样技术发展亟待解决的一些关键问题。

As an effective fast simulation technique in Monte Carlo simulation,Importance Sampling algorithm aims to reduce the variance of a given simulation estimator,so as to reduce the number of simulation samples and the simulation time.In this paper,a short overview of Importance Sampling research is presented,the purpose of research on Importance Sampling and the corresponding basic theory are introduced briefly.Then the Importance Density function is summarized,and some typical application results of Importance Sampling are given.Finally,some crucial issues are proposed for future study.