R package gbev: Gradient boosted regression trees with errors-in-variables. This package performs non-parametric regression when covariates are measured with error. The models are estimated using gradient boosted regression trees. Regression is performed using squared error loss, while binary response regression can be performed using negative log-likelihood loss.
Keywords for this software
References in zbMATH (referenced in 1 article )
Showing result 1 of 1.
- Kriegler, Brian; Berk, Richard: Small area estimation of the homeless in Los Angeles: an application of cost-sensitive stochastic gradient boosting (2010)