Generalized Additive Quantile Regression Model

Quantile Regression with generalized additive models. Calls qgam::qgam() from package qgam.

Form

For the form parameter, a gam formula specific to the Task is required (see example and ?mgcv::formula.gam). If no formula is provided, a fallback formula using all features in the task is used that will make the Learner behave like Linear Quantile Regression. The features specified in the formula need to be the same as columns with col_roles "feature" in the task.

Quantile

The quantile for the Learner, i.e. qu parameter from qgam::qgam(), is set using the value specified in learner$quantiles.

Dictionary

This Learner can be instantiated via lrn():

lrn("regr.qgam")

Meta Information

• Task type: “regr”

• Predict Types: “response”, “se”, “quantiles”

• Feature Types: “logical”, “integer”, “numeric”, “factor”

• Required Packages: mlr3, qgam

Parameters

Id	Type	Default	Levels	Range
form	untyped	-		-
lsig	numeric	-		\((-\infty, \infty)\)
err	numeric	-		\([0, 1]\)
cluster	untyped	NULL		-
multicore	logical	-	TRUE, FALSE	-
ncores	numeric	-		\((-\infty, \infty)\)
paropts	untyped	list()		-
link	untyped	"identity"		-
argGam	untyped	-		-
discrete	logical	FALSE	TRUE, FALSE	-
block.size	integer	1000		\((-\infty, \infty)\)
unconditional	logical	FALSE	TRUE, FALSE	-

References

Fasiolo, Matteo, Wood, N. S, Zaffran, Margaux, Nedellec, Raphael, Goude, Yannig (2017). “Fast Calibrated Additive Quantile Regression.” Journal of the American Statistical Association, 116, 1402–1412. doi:10.1080/01621459.2020.1725521 .

See also

• Dictionary of Learners: mlr3::mlr_learners.

• as.data.table(mlr_learners) for a table of available Learners in the running session (depending on the loaded packages).

• Chapter in the mlr3book: https://mlr3book.mlr-org.com/basics.html#learners

• mlr3learners for a selection of recommended learners.

• mlr3cluster for unsupervised clustering learners.

• mlr3pipelines to combine learners with pre- and postprocessing steps.

• mlr3tuning for tuning of hyperparameters, mlr3tuningspaces for established default tuning spaces.

Author

lona-k

Super classes

mlr3::Learner -> mlr3::LearnerRegr -> LearnerRegrQGam

Methods

Public methods

• LearnerRegrQGam$new()

• LearnerRegrQGam$clone()

Inherited methods

• mlr3::Learner$base_learner()
• mlr3::Learner$configure()
• mlr3::Learner$encapsulate()
• mlr3::Learner$format()
• mlr3::Learner$help()
• mlr3::Learner$predict()
• mlr3::Learner$predict_newdata()
• mlr3::Learner$print()
• mlr3::Learner$reset()
• mlr3::Learner$selected_features()
• mlr3::Learner$train()

---

Method new()

Creates a new instance of this R6 class.

Usage

LearnerRegrQGam$new()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

LearnerRegrQGam$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

# simple example
t = mlr3::tsk("mtcars")
l = mlr3::lrn("regr.qgam")
t$select(c("cyl", "am", "disp", "hp"))
l$param_set$values$form = mpg ~ cyl + am + s(disp) + s(hp)
l$quantiles = 0.25
l$train(t)
#> Estimating learning rate. Each dot corresponds to a loss evaluation. 
#> qu = 0.25...................done 
l$model
#> 
#> Family: elf 
#> Link function: identity 
#> 
#> Formula:
#> mpg ~ cyl + am + s(disp) + s(hp)
#> 
#> Estimated degrees of freedom:
#> 7.79 1.00  total = 11.79 
#> 
#> REML score: 72.30896     
l$predict(t)
#> <PredictionRegr> for 32 observations:
#>  row_ids truth    q0.25 response
#>        1  21.0 20.08142 20.08142
#>        2  21.0 20.08142 20.08142
#>        3  22.8 22.91407 22.91407
#>      ---   ---      ---      ---
#>       30  19.7 18.09929 18.09929
#>       31  15.0 13.42599 13.42599
#>       32  21.4 22.21132 22.21132