Skip to contents

Bayesian exponential smoothing with a nonlinear global trend (LGT/SGT), Student-t errors, and optional heteroscedasticity, fitted via MCMC. The seasonal period is taken from the frequency of the series. Calls Rlgt::rlgt() from package Rlgt.

Dictionary

This mlr3::Learner can be instantiated via the dictionary mlr3::mlr_learners or with the associated sugar function mlr3::lrn():

mlr_learners$get("fcst.rlgt")
lrn("fcst.rlgt")

Meta Information

  • Task type: “fcst”

  • Predict Types: “response”, “quantiles”

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

  • Required Packages: mlr3, mlr3forecast, Rlgt

Parameters

IdTypeDefaultLevelsRange
seasonalityinteger1\([1, \infty)\)
seasonality2integer1\([1, \infty)\)
seasonality.typecharactermultiplicativemultiplicative, generalized-
error.size.methodcharacterstdstd, innov-
level.methodcharacterHWHW, seasAvg, HW_sAvg-
methodcharacterGibbsGibbs, Stan-
homoscedasticlogicalFALSETRUE, FALSE-
controluntypedNULL-
verboselogicalFALSETRUE, FALSE-
NUM_OF_TRIALSinteger2000\([1, \infty)\)

References

Smyl S, Bergmeir C, Wibowo E, Ng TW, Long X, Dokumentov A, Schmidt D (2025). Rlgt: Bayesian Exponential Smoothing Models with Trend Modifications. R package version 0.2-3, https://github.com/cbergmeir/Rlgt.

See also

Other Learner: LearnerFcst, mlr_learners_fcst.adam, mlr_learners_fcst.arfima, mlr_learners_fcst.arima, mlr_learners_fcst.auto_adam, mlr_learners_fcst.auto_arima, mlr_learners_fcst.auto_ces, mlr_learners_fcst.auto_gum, mlr_learners_fcst.auto_msarima, mlr_learners_fcst.auto_ssarima, mlr_learners_fcst.bagged, mlr_learners_fcst.bats, mlr_learners_fcst.ces, mlr_learners_fcst.croston, mlr_learners_fcst.elm, mlr_learners_fcst.es, mlr_learners_fcst.ets, mlr_learners_fcst.gum, mlr_learners_fcst.holt_winters, mlr_learners_fcst.mean, mlr_learners_fcst.mlp, mlr_learners_fcst.msarima, mlr_learners_fcst.nnetar, mlr_learners_fcst.prophet, mlr_learners_fcst.random_walk, mlr_learners_fcst.sma, mlr_learners_fcst.spline, mlr_learners_fcst.ssarima, mlr_learners_fcst.stlm, mlr_learners_fcst.struct_ts, mlr_learners_fcst.tbats, mlr_learners_fcst.theta, mlr_learners_fcst.tscount, mlr_learners_fcst.tslm

Super classes

mlr3::Learner -> mlr3::LearnerRegr -> LearnerFcst -> LearnerFcstForecast -> LearnerFcstRlgt

Methods

Inherited methods


LearnerFcstRlgt$new()

Creates a new instance of this R6 class.

Usage


LearnerFcstRlgt$clone()

The objects of this class are cloneable with this method.

Usage

LearnerFcstRlgt$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

# Define the Learner and set parameter values
learner = lrn("fcst.rlgt")
print(learner)
#> 
#> ── <LearnerFcstRlgt> (fcst.rlgt): Local and Global Trend ───────────────────────
#> • Model: -
#> • Parameters: list()
#> • Packages: mlr3, mlr3forecast, and Rlgt
#> • Predict Types: [response] and quantiles
#> • Feature Types: logical, integer, and numeric
#> • Encapsulation: none (fallback: -)
#> • Properties: exogenous and featureless
#> • Other settings: use_weights = 'error', predict_raw = 'FALSE'

# Define a Task
task = tsk("airpassengers")

# Create train and test set
ids = partition(task)

# Train the learner on the training ids
learner$train(task, row_ids = ids$train)

# Print the model
print(learner$model)
#> $model
#> $n.samples
#> [1] 5000
#> 
#> $sigma2
#> [1] 0.8060902
#> 
#> $xi2
#> [1] 0.8740737
#> 
#> $phi
#> [1] 0.4827586
#> 
#> $chi2
#> [1] 8.437601
#> 
#> $chi2.lambda2
#> [1] 0
#> 
#> $w
#> [1] 0
#> 
#> $alpha
#> [1] 0.8696758
#> 
#> $beta
#> [1] 0.7
#> 
#> $zeta
#> [1] 0.5979935
#> 
#> $rho
#> [1] 0.01724138
#> 
#> $tau
#> [1] 0.3448276
#> 
#> $nu
#> [1] 8.84
#> 
#> $l1
#> [1] 0
#> 
#> $b1
#> [1] 0
#> 
#> $lt
#> [1] 330.698
#> 
#> $bt
#> [1] 0
#> 
#> $et
#> [1] -2.355992
#> 
#> $log.s
#> [1] 0.005258382
#> 
#> $y.on.l
#> [1] 0.01335897
#> 
#> $L
#> [1] 0
#> 
#> $log.s1
#> [1] -0.00432337
#> 
#> $w.s
#> [1] 0
#> 
#> $l2.log.s
#> [1] 0.8105942
#> 
#> $t2.log.s
#> [1] 0.01556474
#> 
#> $s.ix
#> [1] 7
#> 
#> $m
#> [1] 12
#> 
#> $y
#> [1] 200
#> 
#> $mu.hat
#> [1] 205.9295
#> 
#> $method
#> [1] "Gibbs"
#> 
#> $x
#> [1] 200
#> 
#> 
#> $row_ids
#>  [1]  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58
#> [59] 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96
#> 
#> $max_index
#> [1] "1956-12-01"
#> 

# Importance method
if ("importance" %in% learner$properties) print(learner$importance())

# Make predictions for the test rows
predictions = learner$predict(task, row_ids = ids$test)

# Score the predictions
predictions$score()
#> regr.mse 
#> 1106.967