The effect of prevalence and its interaction with sample size on the reliability of species distribution models

By A. Jiménez‐Valverde, Jorge M. Lobo, Joaquín Hortal. 2009

130 citations recorded by the source corpus.

Subfields: Ecological Modeling, Epidemiology, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/e03bb5494a701b90caa88f8ae752244dc9434e1b

How researchers see this paper

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

In Ecological Modeling

In Epidemiology

In Molecular Biology

Citations in the corpus

Geographical patterns in prediction errors of species distribution models (2011, methodology): “the proportion of the number of presences to the number of presences and absences, has been shown to have important effects on the measure of the performance of species distribution models and the selection of the threshold used to convert probability into absences and presences (Jiménez-Valverde et”
Uncertainty associated with survey design in Species Distribution Models (2014, background): “Sample size is also known to have strong effects on SDM predictive accuracy (Hirzel & Guisan, 2002; Reese et al., 2005; Jim enez-Valverde et al., 2009; Chefaoui et al., 2011).”
Stacking species distribution models and adjusting bias by linking them to macroecological models (2014, background): “First, it is well known that issues of sample size plague the estimation of SDMs for rare species (e.g. Jiménez-Valverde et al., 2009).”
Exploring the effects of quantity and location of pseudo-absences and sampling biases on the performance of distribution models with limited point occurrence data (2011, background): “2005), we need to understand how the number of selected pseudo-absence points affects the performance of our models (Jiménez-Valverde et al. 2009).”
Accounting for uncertainty when mapping species distributions: The need for maps of ignorance (2011, background): “Predictive accuracy depends also on the sample size available (e.g. Jiménez-Valverde et al., 2009; Wisz et al., 2008); for small sample sizes, no modelling approach can be robust, and this drawback seriously hinders their usefulness for the immense part of hyperdiverse regions and groups.”
Insights into the area under the receiver operating characteristic curve (AUC) as a discrimination measure in species distribution modelling (2012, methodology): “Sublattices with a prevalence (number of presences/20,000) lower than 0.01 or higher than 0.99 were excluded from the model training step (Jiménez-Valverde et al., 2009a), which left 68 sublattices for model parameterization.”