Variance reduction for diffusions

STOCHASTIC PROCESSES AND THEIR APPLICATIONS

卷:125

Article

Hwang, Chii-Ruey¹  Normand, Raoul¹  Wu, Sheng-Jhih² 

[1] Acad Sinica, Inst Math, Taipei 10617, Taiwan

[2] Suzhou Univ, Sch Math Sci, Ctr Adv Stat & Econometr Res, Suzhou 215006, Peoples R China

关键词: Asymptotic variance;    Rate of convergence;    Diffusion;    Acceleration;    Markov Chain Monte Carlo;   

DOI  :  10.1016/j.spa.2015.03.006

来源: Elsevier

【 摘 要 】

The most common way to sample from a probability distribution is to use Markov Chain Monte Carlo methods. One can find many diffusions with the target distribution as equilibrium measure, so that the state of the diffusion after a long time provides a good sample from that distribution. One naturally wants to choose the best algorithm. One way to do this is to consider a reversible diffusion, and add to it an antisymmetric drift which preserves the invariant measure. We prove that, in general, adding an antisymmetric drift reduces the asymptotic variance, and provide some extensions of this result. (C) 2015 Elsevier B.V. All rights reserved.

【 授权许可】

Free