Skip to contents

Create a new monotonic prior for boosted regressions

Source: R/prior_xgb.R
XGBPrior.Rd

Function to include prior information as monotonic constrain to a extreme gradient descent boosting model engine_xgboost. Monotonic priors enforce directionality in direction of certain variables, however specifying a monotonic constrain does not guarantee that the variable is not regularized out during model fitting.

Usage

XGBPrior(variable, hyper = "increasing", ...)

# S4 method for class 'character,character'
XGBPrior(variable, hyper = "increasing", ...)

Arguments

variable

A character matched against existing predictors or latent effects.

hyper

A character object describing the type of constrain. Available options are 'increasing', 'decreasing', 'convex', 'concave', 'none'.

...

Variables passed on to prior object.

References

  • Chen, T., He, T., Benesty, M., Khotilovich, V., Tang, Y., & Cho, H. (2015). Xgboost: extreme gradient boosting. R package version 0.4-2, 1(4), 1-4.

See also

Prior and GDBPrior.

Other prior: BARTPrior(), BARTPriors(), BREGPrior(), BREGPriors(), GDBPrior(), GDBPriors(), GLMNETPrior(), GLMNETPriors(), INLAPrior(), INLAPriors(), STANPrior(), STANPriors(), XGBPriors(), add_priors(), get_priors(), priors(), rm_priors()

Examples

if (FALSE) { # \dontrun{
 pp <- XGBPrior("forest", "increasing")
} # }