forecaster: forecaster

In jdestefani/ExtendedDFML: Extended Dynamic Factor Machine Learner

Description

Function implementing the forecasting module in the DFML architecture

Usage

forecaster(Z, family = FORECAST_FAMILY, parameters = NULL, h)

Arguments

Z	nxk matrix containing the k time series as columns, each one of length n time steps
family	Forecasting family of method to employ - String among those defined in FORECAST_FAMILY
parameters	Parameters to be passed to the forecaster function - List The method name is passed through the \code{parameters$method} and should be one among those defined in \code{MULTISTEPAHEAD_METHODS}, \code{M4_METHODS}, \code{GRADIENT_BOOSTING_METHODS} For the different methods family, at least the embedding order/model order \code{m} is required. For additional parameters: \itemize{ \item{\code{MULTISTEPAHEAD_METHODS}: }{See \pkg{gbcode::multipleStepAhead} documentation for the role of the different parameters} \item{\code{M4_METHODS}: }{No specific parameters required} \item{\code{GRADIENT_BOOSTING_METHODS}: }{ \itemize{ \item{\code{multistep_method:}{Multistep method to be chosen between \code{direct} and \code{recursive}}} \item{\code{forecasting_method:}{Forecasting method to be chosen between \code{lightgbm} and \code{xgboost}}} } See \pkg{lightgbm} and \pkg{xgboost} documentation for the role of the different parameters} }
h	Forecasting horizon - numeric scalar

Value

List containing:

• Z_hat: h x k matrix containing the h-step ahead forecast for the k input series

• time_forecast: Computational time required to run the forecasting model - numeric scalar

Examples

#See tests/testthat directory on https://github.com/jdestefani/ExtendedDFML

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