作者: M. Hairer , A. M. Stuart , J. Voss , P. Wiberg
DOI: 10.4310/CMS.2005.V3.N4.A8
关键词:
摘要: In many applications it is important to be able sample paths of SDEs conditional on observations various kinds. This paper studies SPDEs which solve such sampling problems. The SPDE may viewed as an infinite dimensional analogue the Langevin SDE used in finite sampling. Here theory developed for conditioned Gaussian processes resulting linear. Applications include Kalman-Bucy filter/smoother. A companion nonlinear case, building linear analysis provided here.