Advanced search
1 file | 4.76 MB Add to list

Spatial regression models for field trials : a comparative study and new ideas

Stijn Hawinkel (UGent) , Sam De Meyer (UGent) and Steven Maere (UGent)
Author
Organization
Project
Abstract
Naturally occurring variability within a study region harbors valuable information on relationships between biological variables. Yet, spatial patterns within these study areas, e.g., in field trials, violate the assumption of independence of observations, setting particular challenges in terms of hypothesis testing, parameter estimation, feature selection, and model evaluation. We evaluate a number of spatial regression methods in a simulation study, including more realistic spatial effects than employed so far. Based on our results, we recommend generalized least squares (GLS) estimation for experimental as well as for observational setups and demonstrate how it can be incorporated into popular regression models for high-dimensional data such as regularized least squares. This new method is available in the BioConductor R-package <jats:italic>pengls</jats:italic>. Inclusion of a spatial error structure improves parameter estimation and predictive model performance in low-dimensional settings and also improves feature selection in high-dimensional settings by reducing “red-shift”: the preferential selection of features with spatial structure. In addition, we argue that the absence of spatial autocorrelation (SAC) in the model residuals should not be taken as a sign of a good fit, since it may result from overfitting the spatial trend. Finally, we confirm our findings in a case study on the prediction of winter wheat yield based on multispectral measurements.
Keywords
Plant Science

Downloads

  • Hawinkel et al. 2022 Frontiers Plant Science 13 858711.pdf
    • full text (Published version)
    • |
    • open access
    • |
    • PDF
    • |
    • 4.76 MB

Citation

Please use this url to cite or link to this publication:

MLA
Hawinkel, Stijn, et al. “Spatial Regression Models for Field Trials : A Comparative Study and New Ideas.” FRONTIERS IN PLANT SCIENCE, vol. 13, 2022, doi:10.3389/fpls.2022.858711.
APA
Hawinkel, S., De Meyer, S., & Maere, S. (2022). Spatial regression models for field trials : a comparative study and new ideas. FRONTIERS IN PLANT SCIENCE, 13. https://doi.org/10.3389/fpls.2022.858711
Chicago author-date
Hawinkel, Stijn, Sam De Meyer, and Steven Maere. 2022. “Spatial Regression Models for Field Trials : A Comparative Study and New Ideas.” FRONTIERS IN PLANT SCIENCE 13. https://doi.org/10.3389/fpls.2022.858711.
Chicago author-date (all authors)
Hawinkel, Stijn, Sam De Meyer, and Steven Maere. 2022. “Spatial Regression Models for Field Trials : A Comparative Study and New Ideas.” FRONTIERS IN PLANT SCIENCE 13. doi:10.3389/fpls.2022.858711.
Vancouver
1.
Hawinkel S, De Meyer S, Maere S. Spatial regression models for field trials : a comparative study and new ideas. FRONTIERS IN PLANT SCIENCE. 2022;13.
IEEE
[1]
S. Hawinkel, S. De Meyer, and S. Maere, “Spatial regression models for field trials : a comparative study and new ideas,” FRONTIERS IN PLANT SCIENCE, vol. 13, 2022.
@article{8748404,
  abstract     = {{Naturally occurring variability within a study region harbors valuable information on relationships between biological variables. Yet, spatial patterns within these study areas, e.g., in field trials, violate the assumption of independence of observations, setting particular challenges in terms of hypothesis testing, parameter estimation, feature selection, and model evaluation. We evaluate a number of spatial regression methods in a simulation study, including more realistic spatial effects than employed so far. Based on our results, we recommend generalized least squares (GLS) estimation for experimental as well as for observational setups and demonstrate how it can be incorporated into popular regression models for high-dimensional data such as regularized least squares. This new method is available in the BioConductor R-package <jats:italic>pengls</jats:italic>. Inclusion of a spatial error structure improves parameter estimation and predictive model performance in low-dimensional settings and also improves feature selection in high-dimensional settings by reducing “red-shift”: the preferential selection of features with spatial structure. In addition, we argue that the absence of spatial autocorrelation (SAC) in the model residuals should not be taken as a sign of a good fit, since it may result from overfitting the spatial trend. Finally, we confirm our findings in a case study on the prediction of winter wheat yield based on multispectral measurements.}},
  articleno    = {{858711}},
  author       = {{Hawinkel, Stijn and De Meyer, Sam and Maere, Steven}},
  issn         = {{1664-462X}},
  journal      = {{FRONTIERS IN PLANT SCIENCE}},
  keywords     = {{Plant Science}},
  language     = {{eng}},
  pages        = {{18}},
  title        = {{Spatial regression models for field trials : a comparative study and new ideas}},
  url          = {{http://doi.org/10.3389/fpls.2022.858711}},
  volume       = {{13}},
  year         = {{2022}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: