Filtering via simulation: auxiliary particle filters

Journal article

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.

Authors: Pitt, M; Shephard, N

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: American Statistical Association Publisher's website
• Journal: Journal of the American Statistical Association
• Volume: 94
• Issue: 446
• Pages: 590-599
• Publication date: 1999-06-01
• ISSN: 0162-1459
• Language: English
• Keywords: Markov chain Monte Carlo; simulation; particle filtering; filtering; sampling/ importance resampling; state space
• Subjects: Economics