Filtering via Simulation: Auxiliary Particle Filters
分析了近期提出的粒子滤波方法在时间序列滤波中的表现，指出该算法因模拟器设计和离散支撑表示而缺乏对异常值的稳健性，并着手解决模拟器设计问题。
Abstract This article analyses the recently suggested particle approach to filtering time series. We suggest that the algorithm is not robust to outliers for two reasons: The design of the simulators and the use of the discrete support to represent the sequentially updating prior distribution. Here we tackle the first of these problems. Key Words: FilteringMarkov chain Monte CarloParticle filterSampling/importance resamplingSimulationState space