Applications of generalized likelihood ratio method to distribution sensitivities and steady-state simulation

作者：Lei, Lei^*; Peng, Yijie; Fu, Michael C.; Hu, Jian-Qiang

来源：Discrete Event Dynamic Systems, 2018, 28(1): 109-125.

DOI：10.1007/s10626-017-0247-8

• Simulation
• Stochastic derivative estimation
• Distribution sensitivity
• Steady-state processes
• Perturbation analysis
• Generalized likelihood ratio method

We provide applications of the generalized likelihood ratio (GLR) method proposed in Peng et al. (2016c) to distribution sensitivity estimation for both finite-horizon and steady-state simulation. Applications on sensitivity of distortion risk measure, gradient-based maximum likelihood estimation, and quantile sensitivity in both finite-horizon and steady-state settings are put together under a single umbrella, and addressed uniformly by the proposed estimator. Empirical comparison of the performance of different methods is presented.

• 出版日期2018-3
• 单位复旦大学