Filtering via Simulation: Auxiliary Particle Filters

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...

Full description

Saved in:
Bibliographic Details
Published inJournal of the American Statistical Association Vol. 94; no. 446; pp. 590 - 599
Main Authors Pitt, Michael K., Shephard, Neil
Format Journal Article
LanguageEnglish
Published Alexandria, VA Taylor & Francis Group 01.06.1999
American Statistical Association
Taylor & Francis Ltd
Subjects
Algorithms
Approximation
Density
Economic models
Evaluation
Exact sciences and technology
Filtering
Filtration
Inference from stochastic processes; time series analysis
Markov chain Monte Carlo
Markov chains
Mathematical economics
Mathematics
Modeling
Non-linear models
Particle density
Particle filter
Probability and statistics
Probability theory and stochastic processes
Quantum statistics
Sampling/importance resampling
Sciences and techniques of general use
Simulation
State space
Statistics
Stochastic processes
Theory and Methods
Time series
Time series models
Outlier
Probability distribution
Filtering
Prior distribution
Algorithm
Time series
Online AccessGet full text

Cover

More Information
Summary: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.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0162-1459
1537-274X
DOI:10.1080/01621459.1999.10474153