Estimating Distribution Sensitivity Using Generalized Likelihood Ratio Method

被引:0
作者
Peng, Yijie [1 ]
Fu, Michael C. [2 ]
Hu, Jian-Qiang [1 ]
机构
[1] Fudan Univ, Sch Management, Shanghai 200433, Peoples R China
[2] Univ Maryland, Robert H Smith Sch Business, College Pk, MD 20740 USA
来源
2016 13TH INTERNATIONAL WORKSHOP ON DISCRETE EVENT SYSTEMS (WODES) | 2016年
关键词
simulation; stochastic derivative estimation; distribution sensitivity; perturbation analysis; likelihood ratio method; ESTIMATING QUANTILE SENSITIVITIES; GRADIENT ESTIMATION; PERTURBATION ANALYSIS; DERIVATIVES; SIMULATION; SYSTEMS;
D O I
暂无
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
We propose a generalized likelihood ratio estimator for the distribution sensitivity in a general framework. Applications on quantile sensitivity, sensitivity of distortion risk measure, and gradient-based maximum likelihood estimation are put together under a single umbrella, and addressed uniformly by the proposed estimator. Numerical experiments substantiate the efficiency of the new method.
引用
收藏
页码:123 / 128
页数:6
相关论文
共 35 条
[21]   INFINITESIMAL AND FINITE PERTURBATION ANALYSIS FOR QUEUING-NETWORKS [J].
HO, YC ;
CAO, X ;
CASSANDRAS, C .
AUTOMATICA, 1983, 19 (04) :439-445
[22]   A NEW APPROACH TO THE ANALYSIS OF DISCRETE EVENT DYNAMIC-SYSTEMS [J].
HO, YC ;
CASSANDRAS, C .
AUTOMATICA, 1983, 19 (02) :149-167
[23]  
Ho Yu-Chi., 1991, Discrete Event Dynamic Systems and Perturbation Analysis
[24]   Pathwise Estimation of Probability Sensitivities Through Terminating or Steady-State Simulations [J].
Hong, L. Jeff ;
Liu, Guangwu .
OPERATIONS RESEARCH, 2010, 58 (02) :357-370
[25]   Estimating Quantile Sensitivities [J].
Hong, L. Jeff .
OPERATIONS RESEARCH, 2009, 57 (01) :118-130
[26]   Technical Note-On Estimating Quantile Sensitivities via Infinitesimal Perturbation Analysis [J].
Jiang, Guangxin ;
Fu, Michael C. .
OPERATIONS RESEARCH, 2015, 63 (02) :435-441
[27]   A UNIFIED VIEW OF THE IPA, SF, AND LR GRADIENT ESTIMATION TECHNIQUES [J].
LECUYER, P .
MANAGEMENT SCIENCE, 1990, 36 (11) :1364-1383
[28]   Kernel Estimation of Quantile Sensitivities [J].
Liu, Guangwu ;
Hong, Liu Jeff .
NAVAL RESEARCH LOGISTICS, 2009, 56 (06) :511-525
[29]  
Peng Y, 2016, OPERATIONS RES UNPUB
[30]  
Peng Y, 2016, WORKING PAPER