sdmpredictors (version 0.2.15)

correlation_groups: Groups layers based on the Pearson correlation

Description

correlation_groups returns groups of layer codes such as each layer from one group has an absolute Pearson product-moment correlation coefficient (Pearson's r) that is smaller than the maximum_correlation (default 0.7) with each variable in any other group. The correlation values of quadratic layers are used for creating the groups but only non quadratic layer codes are returned.

Usage

correlation_groups(layers_correlation, max_correlation=0.7)

Value

A list of vectors with each vector containing the layer codes of one correlation group.

Arguments

layers_correlation

matrix or dataframe. A square matrix with the layers correlations you want to group.

max_correlation

number. The maximum correlation 2 layers may have before they are put in the same correlation group.

References

Dormann, C. F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carre, G., ... Lautenbach, S. (2013). Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography, 36(1), 027-046. doi:10.1111/j.1600-0587.2012.07348.x Barbet-Massin, M. & Jetz, W. (2014). A 40-year, continent-wide, multispecies assessment of relevant climate predictors for species distribution modelling. Diversity and Distributions, 20(11), 1285-1295. doi:10.1111/ddi.12229

See Also

layers_correlation list_layers layer_stats

Examples

corr <- layers_correlation(c("BO_calcite", "BO_damin", "MS_bathy_5m"))
print(corr)
print(correlation_groups(corr, max_correlation=0.6))