Event Abstract

Back to Event

An implementation of Geographical and Temporal Weighted Regression in modeling occurrences of Porcine Reproductive and Respiratory Syndrome in the US swine industry

Keith Sollers1*, Xin Liu2 and Beatriz Martínez-López3
  • 1 University of California, Davis, United States
  • 2 Department of Computer Science, University of California, Davis, United States
  • 3 Center for Animal Disease Modeling and Surveillance, School of Veterinary Medicine, University of California, Davis, United States

Geographical and Temporal Weighted Regression (GTWR) has emerged as an effective extension of Geographically Weighted Regression (GWR) for modeling nonstationary spatiotemporal processes such as those that commonly arise from epidemiological study (1). Briefly, GTWR combines local linear regressions in both temporal and spatial dimensions into a global model through the use of a spatiotemporal kernel function. Bandwidth optimization first in the temporal dimension and then the spatial dimension produces a cohesive model across the entire space of interest that accounts for historical temporal variation. The resulting GTWR model can thus be used not only to model current data but also as a more reliable predictor of future events (e.g. outbreaks). Specifically, in this study we apply GTWR to evaluate the occurrence of swine diseases in the US swine industry. We illustrate the method using animal movement network and diagnostic time-series data that arise from the ongoing surveillance activities of Porcine Reproductive and Respiratory Syndrome (PRRS). Using a domain-specific spatiotemporal kernel function, the resulting optimal GTWR model is shown to improve empirically on standard GWR methods. Finally, model residuals are analyzed and an assessment of the power of the model in predicting future PRRS outbreaks is conducted. Our results show that the optimal GTWR model improves empirically on standard GWR methods. Results reveal that GTWR is a valuable tool to gather insights in the spatiotemporal patterns of PRRS and will likely serve for any other swine disease or diseases affecting other livestock species. This approach is relatively easy to implement with animal health data that is routinely collected to inform risk-based surveillance and control programs, which certainly will benefit the implementation of more cost-effective prevention and mitigation strategies in the livestock industry.

Acknowledgements

This project was partially funded by the NSF BIGDATA:IA Award #1838207 and by the Innovator Fellowship from the Innovation Institute for Food and Health. Authors would like to acknowledge swine industry collaborators for the provision of data.

References

1. Fotheringham, Crespo, & Yao (2015). "Geographical and Temporal Weighted Regression (GTWR)". Geographical Analysis, 1-22.

Keywords: GTWR, Epidemiology, PRRs, Spatio - temporal analysis, Swine industry

Conference: GeoVet 2019. Novel spatio-temporal approaches in the era of Big Data, Davis, United States, 8 Oct - 10 Oct, 2019.

Presentation Type: Student oral presentation

Topic: Spatio-temporal surveillance and modeling approaches

Citation: Sollers K, Liu X and Martínez-López B (2019). An implementation of Geographical and Temporal Weighted Regression in modeling occurrences of Porcine Reproductive and Respiratory Syndrome in the US swine industry. Front. Vet. Sci. Conference Abstract: GeoVet 2019. Novel spatio-temporal approaches in the era of Big Data. doi: 10.3389/conf.fvets.2019.05.00012

Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.

The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.

Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.

For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.

Received: 21 Jun 2019; Published Online: 27 Sep 2019.

* Correspondence: Mx. Keith Sollers, University of California, Davis, Davis, United States, keith.sollers@gmail.com