Multivariate epidemic count time series model.
An infectious disease spreads not only over a single population or community but also across multiple and heterogeneous communities. Moreover, its transmissibility varies over time because of various factors such as seasonality and epidemic control, which results in strongly nonstationary behavior....
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2023-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0287389 |
| _version_ | 1797799827902300160 |
|---|---|
| author | Shinsuke Koyama |
| author_facet | Shinsuke Koyama |
| author_sort | Shinsuke Koyama |
| collection | DOAJ |
| description | An infectious disease spreads not only over a single population or community but also across multiple and heterogeneous communities. Moreover, its transmissibility varies over time because of various factors such as seasonality and epidemic control, which results in strongly nonstationary behavior. In conventional methods for assessing transmissibility trends or changes, univariate time-varying reproduction numbers are calculated without taking into account transmission across multiple communities. In this paper, we propose a multivariate-count time series model for epidemics. We also propose a statistical method for estimating the transmission of infections across multiple communities and the time-varying reproduction numbers of each community simultaneously from a multivariate time series of case counts. We apply our method to incidence data for the novel coronavirus disease 2019 (COVID-19) pandemic to reveal the spatiotemporal heterogeneity of the epidemic process. |
| first_indexed | 2024-03-13T04:25:50Z |
| format | Article |
| id | doaj.art-5e960cb3a1fd422da98624018a131518 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-03-13T04:25:50Z |
| publishDate | 2023-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-5e960cb3a1fd422da98624018a1315182023-06-20T05:30:57ZengPublic Library of Science (PLoS)PLoS ONE1932-62032023-01-01186e028738910.1371/journal.pone.0287389Multivariate epidemic count time series model.Shinsuke KoyamaAn infectious disease spreads not only over a single population or community but also across multiple and heterogeneous communities. Moreover, its transmissibility varies over time because of various factors such as seasonality and epidemic control, which results in strongly nonstationary behavior. In conventional methods for assessing transmissibility trends or changes, univariate time-varying reproduction numbers are calculated without taking into account transmission across multiple communities. In this paper, we propose a multivariate-count time series model for epidemics. We also propose a statistical method for estimating the transmission of infections across multiple communities and the time-varying reproduction numbers of each community simultaneously from a multivariate time series of case counts. We apply our method to incidence data for the novel coronavirus disease 2019 (COVID-19) pandemic to reveal the spatiotemporal heterogeneity of the epidemic process.https://doi.org/10.1371/journal.pone.0287389 |
| spellingShingle | Shinsuke Koyama Multivariate epidemic count time series model. PLoS ONE |
| title | Multivariate epidemic count time series model. |
| title_full | Multivariate epidemic count time series model. |
| title_fullStr | Multivariate epidemic count time series model. |
| title_full_unstemmed | Multivariate epidemic count time series model. |
| title_short | Multivariate epidemic count time series model. |
| title_sort | multivariate epidemic count time series model |
| url | https://doi.org/10.1371/journal.pone.0287389 |
| work_keys_str_mv | AT shinsukekoyama multivariateepidemiccounttimeseriesmodel |