R/mrf_multi_step_forecast.R

In Quirinms/MRFR: Multiresolution Forecasting

mrf_multi_step_forecast <- function(UnivariateData, Horizon, Aggregation,
                                    CoefficientCombination=NULL,
                                    Method = "r",
                                    Threshold = "hard",
                                    Lambda = 0.05){
  # DESCRIPTION
  # Computes multi-step forecasts with the multiresolution method.
  #
  # INPUT
  # UnivariateData[1:n]                   Numerical vector with n time series
  #                                       values.
  # Horizon          Number indicating horizon for forecast from 1 to horizon.
  # Aggregation[1:Scales]    Numerical vector carrying numbers whose index is
  #                          associated with the wavelet level. The numbers
  #                          indicate the number of time in points used for
  #                          aggregation from the original time series.
  #
  # OPTIONAL
  # CoefficientCombination[1:Scales+1]    Numerical vector with numbers which
  #                                       are associated with wavelet levels.
  #                                       The last number is associated with the
  #                                       smooth level. Each number determines
  #                                       the number of coefficient used per
  #                                       level. The selection follows a
  #                                       specific scheme.
  # Threshold                Character indicating if Thresholding is done on the
  #                          wavelet decomposition or not.
  #                          Default: Threshold="hard". Possible entries:
  #                          Threshold = "hard" for hard thresholding.
  #                          Threshold = "soft" for soft thresholding.
  #                          Any other input indicates no thresholding.
  # Lambda                   Numeric value indicating the threshold for
  #                          computing a hard or soft threshold on the wavelet
  #                          decomposition.
  # Method           String indicating which method to use
  #                  Available methods: 'r'  = Autoregression
  #                                     'nn' = Neural Network
  #                                     'elm' = Extreme Learning Machine
  #                                     'nnetar' = forecast::nnetar
  #
  # OUTPUT
  # multistep[1:Horizon]    Numerical vector with forecast of horizon according
  #                         to its index.
  #
  # Author: QS, 02/2021
  if(!is.vector(UnivariateData)){
    message("Data must be of type vector")
    return()
  }
  if(!is.double(Horizon)){
    message("steps must be of type double")
    return()
  }
  if(!is.vector(Aggregation)){
    message("agg_per_lvl must be of type vector")
    return()
  }
  if(!is.character(Method)){
    message("method must be of type character")
    return()
  }
  if(Method %in% c("r", "nn")){
    if(!is.vector(CoefficientCombination)){
      message("ccps must be of type vector")
      return()
    }
  }
  if(is.null(CoefficientCombination) && (Method %in% c("r", "nn"))){
    stop("CoefficientCombination must be given for all methods except 'elm' and 'nnetar'.")
    return()
  }

  multistep = c()
  # recursive multi-steps
  if(Method %in% c("r", "nn")){
    for(i in 1:Horizon){
      forecast = 0
      forecast = mrf_one_step_forecast(UnivariateData=UnivariateData,
                                       CoefficientCombination=CoefficientCombination,
                                       Aggregation=Aggregation,
                                       Method=Method,
                                       Threshold=Threshold,
                                       Lambda=Lambda)
      multistep = c(multistep, forecast)
      UnivariateData = as.vector(unlist(c(UnivariateData, forecast)))

    }
  # non-recursive multi-steps
  }else if(Method == "elm"){
    multistep = mrf_elm_forecast(UnivariateData, Horizon, Aggregation,
                                 Threshold, Lambda)
  }else{
    multistep = mrf_nnetar_forecast(UnivariateData, Horizon, Aggregation,
                                    Threshold, Lambda)
  }
  multistep = array(unlist(multistep))
  # Cap forecasts exceeding certain limits
  max_val = max(UnivariateData)
  upper_limit = 0
  if(max_val > 0){
    upper_limit = 1.3*max(UnivariateData)
  }else{
    upper_limit = 0.7*max(UnivariateData)
  }
  min_val = min(UnivariateData)
  lower_limit = 0
  if(min_val > 0){
    lower_limit = 0.7*min(UnivariateData)
  }else{
    lower_limit = 1.3*min(UnivariateData)
  }
  for(i in 1:Horizon){
    forecast = multistep[i]
    if(forecast > upper_limit){
      forecast = upper_limit
    }
    if(forecast < lower_limit){
      forecast = lower_limit
    }
    multistep[i] = forecast
  }
  return(multistep)
}


#
#
#
#
#