gradientBoostingForecaster: gradientBoostingForecaster

In jdestefani/ExtendedDFML: Extended Dynamic Factor Machine Learner

Description

gradientBoostingForecaster

Usage

gradientBoostingForecaster(
  ts,
  embedding,
  horizon,
  delay = 0,
  multistep_method = c("recursive", "direct"),
  forecasting_method = c("lightgbm", "xgboost"),
  forecasting_params = NULL
)

Arguments

ts	Input time series as a vector
embedding	Embedding order
horizon	Forecasting horizon
delay	Delay for forecasting
multistep_method	Multistep method (to be chosen among "recursive" and "direct")
forecasting_method	Gradient boosting forecasting method (to be chosen among "lightgbm" and "xgboost")
forecasting_params	Parameters to be passed to the forecaster function - List multistep_method:Multistep method to be chosen between direct and recursive forecasting_method:Forecasting method to be chosen between lightgbm and xgboost See \pkg{lightgbm} and \pkg{xgboost} documentation for the role of the different parameters

Value

h-step forecast of ts using the chosen multistep_method and forecasting method, with the given embedding order

jdestefani/ExtendedDFML documentation built on Dec. 20, 2021, 10:04 p.m.