CRAN

Version 1.1.3, available on CRAN

spatialRF: Easy Spatial Modeling with Random Forest

Automatic generation and selection of spatial predictors for spatial regression with Random Forest. Spatial predictors are surrogates of variables driving the spatial structure of a response variable. The package offers two methods to generate spatial predictors from a distance matrix among training cases: 1) Moran's Eigenvector Maps (MEMs; Dray, Legendre, and Peres-Neto 2006 DOI:10.1016/j.ecolmodel.2006.02.015): computed as the eigenvectors of a weighted matrix of distances; 2) RFsp (Hengl et al. DOI:10.7717/peerj.5518): columns of the distance matrix used as spatial predictors. Spatial predictors help minimize the spatial autocorrelation of the model residuals and facilitate an honest assessment of the importance scores of the non-spatial predictors. Additionally, functions to reduce multicollinearity, identify relevant variable interactions, tune random forest hyperparameters, assess model transferability via spatial cross-validation, and explore model results via partial dependence curves and interaction surfaces are included in the package. The modelling functions are built around the highly efficient 'ranger' package (Wright and Ziegler 2017 DOI:10.18637/jss.v077.i01).

spatialRF: Easy Spatial Regression with Random Forest

Spatial explanatory regression with Random Forest via generation and selection of spatial predictors. Spatial predictors are surrogates of variables responsible for the spatial structure of the data, generated from the distance matrix via Moran's Eigenvector Maps (Dray, Legendre, and Peres-Neto 2006 DOI:10.1016/j.ecolmodel.2006.02.015) or by including the complete distance matrix as a spatial component in a model (Hengl et al. DOI:10.7717/peerj.5518). Spatial predictors help minimize the spatial autocorrelation of the model residuals and facilitate an honest assessment of the importance scores of the non-spatial predictors. Among others, the package includes functions to reduce multicollinearity, identify relevant variable interactions, tune random forest hyperparameters, assess model transferability via spatial cross-validation, and explore model results via partial dependence curves and interaction surfaces. The modelling functions are built around the highly efficient 'ranger' package (Wright and Ziegler 2017 DOI:10.18637/jss.v077.i01) and are designed to run in parallel in a single machine or a Beowulf cluster.