Many zeros does not mean zero inflation: comparing the goodness‐of‐fit of parametric models to multivariate abundance data

By David I. Warton. 2005

231 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Ecological Modeling, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/6e2c9c26626dccb9bd11c1eaed06a21df2912eb3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Ecological Modeling

In Nature and Landscape Conservation

Citations in the corpus

Distance‐based multivariate analyses confound location and dispersion effects (2012, methodology): “4c) can be addressed using the simple approach of fitting separate models to each taxon, then summing the test statistics, and using resampling to assess the significance of this multivariate test statistic (Warton 2011). This ‘sum-of-LR’ statistic is analogous to the ‘sum-of-F’ approach based on an”
Comparison of five modelling techniques to predict the spatial distribution and abundance of seabirds (2012, methodology): “Such data are difficult to incorporate into standard parametric models (Martin et al., 2005; Sileshi et al., 2009; Warton, 2005).”
The excess-zero problem in soil animal count data and choice of appropriate models for statistical inference (2008, methodology): “Sampling of many taxa simultaneously is not usually limited to locations where all taxa might occur, hence multivariate abundance data are naturally expected to contain more frequent zeros (Sileshi, 2006; Warton, 2005).”
Modeling zero-inflated count data with glmmTMB (2017, result): “These results agree with earlier work showing that the negative binomial distribution is often the best fit for abundance data (Warton, 2005).”
glmmTMB Balances Speed and Flexibility Among Packages for Zero-inflated Generalized Linear Mixed Modeling (2017, background): “It did not include zero-inﬂation, as is common in abundance models (Warton, 2005).”
Penalized Normal Likelihood and Ridge Regularization of Correlation and Covariance Matrices (2008, methodology): “Although this approach has been shown to do a surprisingly good job of modeling the mean–variance relationship of this type of data set (Warton 2005), it nevertheless is a rather crude modeling approach and thus can be considered a first attempt that may be improved on in later work.”