Auto.Arima Forecast Learner

Auto ARIMA model Calls forecast::auto.arima from package forecast.

Dictionary

This Learner can be instantiated via the dictionary mlr_learners or with the associated sugar function lrn():

mlr_learners$get("forecast.auto_arima")
lrn("forecast.auto_arima")

Meta Information

• Task type: “forecast”

• Predict Types: “response”, “se”

• Feature Types: “numeric”

• Required Packages: mlr3, forecast

Parameters

Id	Type	Default	Levels	Range
d	integer	NA		\([0, \infty)\)
D	integer	NA		\([0, \infty)\)
max.q	integer	5		\([0, \infty)\)
max.p	integer	5		\([0, \infty)\)
max.P	integer	2		\([0, \infty)\)
max.Q	integer	2		\([0, \infty)\)
max.order	integer	5		\([0, \infty)\)
max.d	integer	2		\([0, \infty)\)
max.D	integer	1		\([0, \infty)\)
start.p	integer	2		\([0, \infty)\)
start.q	integer	2		\([0, \infty)\)
start.P	integer	2		\([0, \infty)\)
start.Q	integer	2		\([0, \infty)\)
stepwise	logical	FALSE	TRUE, FALSE	-
allowdrift	logical	TRUE	TRUE, FALSE	-
seasonal	logical	FALSE	TRUE, FALSE	-

See also

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

• Package mlr3learners for a solid collection of essential learners.

• Package mlr3extralearners for more learners.

• Dictionary of Learners: mlr_learners

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

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

• Package mlr3viz for some generic visualizations.

• Extension packages for additional task types:

  • mlr3proba for probabilistic supervised regression and survival analysis.

  • mlr3cluster for unsupervised clustering.

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

Other Learner: LearnerForecast, mlr_learners_regr.VAR, mlr_learners_regr.arima, mlr_learners_regr.average

Super classes

mlr3::Learner -> mlr3temporal::LearnerForecast -> LearnerRegrForecastAutoArima

Methods

Public methods

• LearnerRegrForecastAutoArima$new()

• LearnerRegrForecastAutoArima$forecast()

• LearnerRegrForecastAutoArima$clone()

Inherited methods

• mlr3::Learner$base_learner()
• mlr3::Learner$format()
• mlr3::Learner$help()
• mlr3::Learner$predict_newdata()
• mlr3::Learner$print()
• mlr3::Learner$reset()
• mlr3temporal::LearnerForecast$fitted_values()
• mlr3temporal::LearnerForecast$predict()
• mlr3temporal::LearnerForecast$train()

---

Method new()

Creates a new instance of this R6 class.

Usage

LearnerRegrForecastAutoArima$new()

---

Method forecast()

Returns forecasts after the last training instance.

Usage

LearnerRegrForecastAutoArima$forecast(h = 10, task, new_data = NULL)

Arguments

h

(numeric(1))
Number of steps ahead to forecast. Default is 10.

task

(Task).

new_data

(data.frame())
New data to predict on.

Returns

Prediction.

---

Method clone()

The objects of this class are cloneable with this method.

Usage

LearnerRegrForecastAutoArima$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

learner = mlr3::lrn("forecast.auto_arima")
print(learner)
#> <LearnerRegrForecastAutoArima:forecast.auto_arima>
#> * Model: -
#> * Parameters: list()
#> * Packages: mlr3, forecast
#> * Predict Types:  [response], se
#> * Feature Types: numeric
#> * Properties: exogenous, missings, univariate

# available parameters:
learner$param_set$ids()
#>  [1] "d"          "D"          "max.q"      "max.p"      "max.P"     
#>  [6] "max.Q"      "max.order"  "max.d"      "max.D"      "start.p"   
#> [11] "start.q"    "start.P"    "start.Q"    "stepwise"   "allowdrift"
#> [16] "seasonal"