Modelling multi‐scale, state‐switching functional data with hidden Markov models
Data sets composed of sequences of curves sampled at high frequencies in time are increasingly common in practice, but they can exhibit complicated dependence structures that cannot be modelled using common methods in functional data analysis. We detail a hierarchical approach that treats the curves...
Published in: | Canadian Journal of Statistics |
---|---|
Main Authors: | , , , , , |
Other Authors: | , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Wiley
2021
|
Subjects: | |
Online Access: | http://dx.doi.org/10.1002/cjs.11673 https://onlinelibrary.wiley.com/doi/pdf/10.1002/cjs.11673 https://onlinelibrary.wiley.com/doi/full-xml/10.1002/cjs.11673 |
id |
crwiley:10.1002/cjs.11673 |
---|---|
record_format |
openpolar |
spelling |
crwiley:10.1002/cjs.11673 2024-06-23T07:54:22+00:00 Modelling multi‐scale, state‐switching functional data with hidden Markov models Sidrow, Evan Heckman, Nancy Fortune, Sarah M. E. Trites, Andrew W. Murphy, Ian Auger‐Méthé, Marie Natural Sciences and Engineering Research Council of Canada Fisheries and Oceans Canada 2021 http://dx.doi.org/10.1002/cjs.11673 https://onlinelibrary.wiley.com/doi/pdf/10.1002/cjs.11673 https://onlinelibrary.wiley.com/doi/full-xml/10.1002/cjs.11673 en eng Wiley http://onlinelibrary.wiley.com/termsAndConditions#vor Canadian Journal of Statistics volume 50, issue 1, page 327-356 ISSN 0319-5724 1708-945X journal-article 2021 crwiley https://doi.org/10.1002/cjs.11673 2024-06-11T04:41:58Z Data sets composed of sequences of curves sampled at high frequencies in time are increasingly common in practice, but they can exhibit complicated dependence structures that cannot be modelled using common methods in functional data analysis. We detail a hierarchical approach that treats the curves as observations from a hidden Markov model. The distribution of each curve is then defined by another fine‐scale model that may involve autoregression and require data transformations using moving‐window summary statistics or Fourier analysis. This approach is broadly applicable to sequences of curves exhibiting intricate dependence structures. As a case study, we use this framework to model the fine‐scale kinematic movements of a northern resident killer whale ( Orcinus orca ) off the western coast of Canada. Through simulations, we show that our model produces more interpretable state estimation and more accurate parameter estimates compared to existing methods. Article in Journal/Newspaper Killer Whale Orca Orcinus orca Killer whale Wiley Online Library Canada Canadian Journal of Statistics 50 1 327 356 |
institution |
Open Polar |
collection |
Wiley Online Library |
op_collection_id |
crwiley |
language |
English |
description |
Data sets composed of sequences of curves sampled at high frequencies in time are increasingly common in practice, but they can exhibit complicated dependence structures that cannot be modelled using common methods in functional data analysis. We detail a hierarchical approach that treats the curves as observations from a hidden Markov model. The distribution of each curve is then defined by another fine‐scale model that may involve autoregression and require data transformations using moving‐window summary statistics or Fourier analysis. This approach is broadly applicable to sequences of curves exhibiting intricate dependence structures. As a case study, we use this framework to model the fine‐scale kinematic movements of a northern resident killer whale ( Orcinus orca ) off the western coast of Canada. Through simulations, we show that our model produces more interpretable state estimation and more accurate parameter estimates compared to existing methods. |
author2 |
Natural Sciences and Engineering Research Council of Canada Fisheries and Oceans Canada |
format |
Article in Journal/Newspaper |
author |
Sidrow, Evan Heckman, Nancy Fortune, Sarah M. E. Trites, Andrew W. Murphy, Ian Auger‐Méthé, Marie |
spellingShingle |
Sidrow, Evan Heckman, Nancy Fortune, Sarah M. E. Trites, Andrew W. Murphy, Ian Auger‐Méthé, Marie Modelling multi‐scale, state‐switching functional data with hidden Markov models |
author_facet |
Sidrow, Evan Heckman, Nancy Fortune, Sarah M. E. Trites, Andrew W. Murphy, Ian Auger‐Méthé, Marie |
author_sort |
Sidrow, Evan |
title |
Modelling multi‐scale, state‐switching functional data with hidden Markov models |
title_short |
Modelling multi‐scale, state‐switching functional data with hidden Markov models |
title_full |
Modelling multi‐scale, state‐switching functional data with hidden Markov models |
title_fullStr |
Modelling multi‐scale, state‐switching functional data with hidden Markov models |
title_full_unstemmed |
Modelling multi‐scale, state‐switching functional data with hidden Markov models |
title_sort |
modelling multi‐scale, state‐switching functional data with hidden markov models |
publisher |
Wiley |
publishDate |
2021 |
url |
http://dx.doi.org/10.1002/cjs.11673 https://onlinelibrary.wiley.com/doi/pdf/10.1002/cjs.11673 https://onlinelibrary.wiley.com/doi/full-xml/10.1002/cjs.11673 |
geographic |
Canada |
geographic_facet |
Canada |
genre |
Killer Whale Orca Orcinus orca Killer whale |
genre_facet |
Killer Whale Orca Orcinus orca Killer whale |
op_source |
Canadian Journal of Statistics volume 50, issue 1, page 327-356 ISSN 0319-5724 1708-945X |
op_rights |
http://onlinelibrary.wiley.com/termsAndConditions#vor |
op_doi |
https://doi.org/10.1002/cjs.11673 |
container_title |
Canadian Journal of Statistics |
container_volume |
50 |
container_issue |
1 |
container_start_page |
327 |
op_container_end_page |
356 |
_version_ |
1802646505880813568 |