Extending Ordinal Regression with a Latent Zero-Augmented Beta Distribution
Ecological abundance data are often recorded on an ordinal scale in which the lowest category represents species absence. One common example is when plant species cover is visually assessed within bounded quadrats and then assigned to pre-defined cover class categories. We present an ordinal beta hurdle model that directly models ordinal category probabilities with a biologically realistic beta-distributed latent variable. A hurdle-at-zero model allows ecologists to explore distribution (absence) and abundance processes in an integrated framework. This provides an alternative to cumulative link models when data are inconsistent with the assumption that the odds of moving into a higher category are the same for all categories (proportional odds). Graphical tools and a deviance information criterion were developed to assess whether a hurdle-at-zero model should be used for inferences rather than standard ordinal methods. Hurdle-at-zero and non-hurdle ordinal models fit to vegetation cover class data produced substantially different conclusions. The ordinal beta hurdle model yielded more precise parameter estimates than cumulative logit models, although out-of-sample predictions were similar. The ordinal beta hurdle model provides inferences directly on the latent biological variable of interest, percent cover, and supports exploration of more realistic ecological patterns and processes through the hurdle-at-zero or two-part specification. We provide JAGS code as an on-line supplement. Supplementary materials accompanying this paper appear on-line.
KeywordsBeta regression Cumulative link model Grouped continuous Hurdle model Midpoint regression Non-proportional odds Plant abundance Proportional odds model
We thank Dr. Megan D. Higgs for early discussions on this work and her assistance with WinBUGS code for clipping latent distributions. Dr. Brian Gray provided encouragement and interest in this work and we are appreciative. We also thank Dr. Andrew Hoegh, two anonymous reviewers’, and our associate editor’s comments and suggestion for revising our paper. The work by K. M. Irvine was funded through an Interagency Agreement P12PG70586 with the National Park Service. T. J. Rodhouse was funded by Upper Columbia Basin Network Inventory and Monitoring Program of the National Park Service. I. N. Keren’s participation was secured by an interagency agreement with Montana State’s Institute on Ecosystems with funding by North Central Climate Science Center. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.
- Andrewartha, H. G., and Birch, L. C. (1954), The distribution and abundance of animals. University of Chicago Press, Chicago, Illinois, USA.Google Scholar
- Bonham, C. D. (1989), Measurements for terrestrial vegetation John Wiley and Sons, New York, NY.Google Scholar
- Braun-Blanquet, J. (1932), Plant sociology. The study of plant communities. McGraw-Hill, New York, NY, US.Google Scholar
- Christensen, R. H. B. (2014), Ordinal:Regression models for ordinal data. R package version 2014.12-22. Available at http://cran.r-project.org/web/packages/ordinal/index.html.
- Daubenmire, R. F. (1959), A canopy-coverage method. Northwest Science, 33, 43–64.Google Scholar
- Duff, T. J., Bell, T. L., and York, A. (2011), Patterns of plant abundances in natural systems: is there value in modelling both species abundance and distribution?. Australian Journal of Botany, 59, 719–733.Google Scholar
- Eskelson, N. I., Madsen, L., Hagar, J. C., and Temesgen, H. (2011), Estimating riparian understory vegetation cover with beta regression and copula models. Forest Science, 57, 212–221.Google Scholar
- Esposito, D. M., Shanahan, E., and Rodhouse, T. J. (2016), UCBN and GRYN Sagebrush Steppe Vegetation Monitoring: Double Observer Study 2015. John Day Fossil Beds National Monument-Clarno Unit and City of Rocks National Reserve. Natural Resource Reporting Series NPS/UCBN/NRR-2016/1052. National Park Service, Fort Collins, Colorado.Google Scholar
- Fahrmeier, L, and Tutz, G. (2001), Multivariate statistical modelling based on generalized linear models. Springer, Berlin.Google Scholar
- Gruen, B., Kosmidis, I., and Zeileis, A. (2012), Extended Beta Regression in R: Shaken, Stirred, Mixed, and Partitioned. Journal of Statistical Software, 48, 1–25.Google Scholar
- Ishwaran, H. (2000), Univariate and multirater ordinal cumulative link regression with covariate specific cutpoints. The Canadian Journal of Statistics, 28, 715–730.Google Scholar
- Mackenzie, D. I., Nichols, J. D., Royle, J. A., Pollock, K. H., Bailey, L. L., and Hines, J. E. (2006), Occupancy estimation and modeling:inferring patterns and dynamics of species occurrence. Elsevier Academic Press, Burlington, MA, USA.Google Scholar
- Miller, R. F., Chambers, J. C., Pyke, D. A., Pierson, F. B., and Williams, C. J. (2013), A review of fire effects on vegetation and soils in the Great Basin region: response and ecological site characteristics. RMRS GTR-308. USDA Forest Service, Rocky Mountain Research Station, Fort Collins, Colorado, USA.Google Scholar
- Plummer, M. (2003), JAGS: A program for analysis of Bayesian graphical models using Gibbs sampling. In Proceedings of the 3rd International Workshop on Distributed Statistical Computing (DSC 2003), K. Hornik, F. Leisch, and A. Zeileis (eds.) Vienna, Austria. Available at: http://www.ci.tuwien.ac.at/Conferences/DSC-2003/
- Plummer, M. (2015), JAGS Version 4.0.0 user manual. Available at https://sourceforge.net/projects/mcmc-jags/files/Manuals/4.x/.
- Rodhouse, T. J., Irvine, K. M., Sheley, R. L., Smith, B. S., Hoh, S., Esposito, D. M., and Mata-Gonzalez, R. (2014), Predicting foundation bunchgrass species abundances: model-assisted decision-making in protected-area sagebrush-steppe. Ecosphere, 5, art208.Google Scholar
- Schabenberger, O. (1995), The use of ordinal response methodology in Forestry. Forest Science, 41, 321–336.Google Scholar
- Spiegelhalter, D., Best, N., Carlin, B., and van der Linde, A. (2002), Bayesian measures of model complexity and fit (with discussion). Journal of the Royal Statistical Society B, 64, 583–639.Google Scholar
- Yeo, J. J., Rodhouse, T. J., Dicus, G. H., Irvine, K. M., and Garrett, L. K. (2009), Sagebrush steppe vegetation monitoring protocol. Upper Columbia Basin Network. Version 1.0. Natural Resource Report NPS/UCBN/NRR–2009/142. National Park Service, Fort Collins, CO, USA.Google Scholar