A model building procedure to build parsimonious geoadditive model from a large number of covariates. Continuous, binary and ordered categorical responses are supported. The model building is based on component wise gradient boosting with linear effects, smoothing splines and a smooth spatial surface to model spatial autocorrelation. The resulting covariate set after gradient boosting is further reduced through backward elimination and aggregation of factor levels. The package provides a model based bootstrap method to simulate prediction intervals for point predictions. A test data set of a soil mapping case study in Berne (Switzerland) is provided.
| Version: | 0.1-2 |
| Depends: | R (≥ 2.14.0) |
| Imports: | mboost, mgcv, grpreg, MASS |
| Published: | 2017-07-23 |
| Author: | Madlene Nussbaum [cre, aut], Andreas Papritz [ths] |
| Maintainer: | Madlene Nussbaum <madlene.nussbaum at env.ethz.ch> |
| License: | GPL-2 | GPL-3 [expanded from: GPL (≥ 2)] |
| NeedsCompilation: | no |
| Materials: | NEWS |
| CRAN checks: | geoGAM results |
| Reference manual: | geoGAM.pdf |
| Package source: | geoGAM_0.1-2.tar.gz |
| Windows binaries: | r-devel: geoGAM_0.1-2.zip, r-release: geoGAM_0.1-2.zip, r-oldrel: geoGAM_0.1-2.zip |
| macOS binaries: | r-release: geoGAM_0.1-2.tgz, r-oldrel: geoGAM_0.1-2.tgz |
| Old sources: | geoGAM archive |
Please use the canonical form https://CRAN.R-project.org/package=geoGAM to link to this page.