Publication:
Detection of outlier patches in autoregressive time series

Loading...
Thumbnail Image
Identifiers
Publication date
1998-02
Defense date
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Google Scholar
Export
Research Projects
Organizational Units
Journal Issue
Abstract
This paper proposed a procedure to identify patches of outliers in an autoregressive process. The procedure is an improvement over the existing outlier detection methods via Gibbs sampling. It identifies the beginning and end of possible outlier patches using the existing Gibbs sampling, then carries out and adaptive procedure with block interpolation to handle patches of outliers. Empirical and simulated examples show that the proposed procedure is effective in handling masking and swamping effects caused by multiple outliers. The real example also shows that the standard Gibbs sampling to outlier detection may encounter severe masking and swamping effects in practice.
Description
Keywords
Gibbs sampler, Time series, Multiple outliers, Sequential learning
Bibliographic citation