add_univariate_forecast_model: Add univariate forecast model

In ing-bank/tsforecast: Time Series Forecasting Pipeline

Description

add_univariate_forecast_model is a function to add a single univariate forecast model to a (named) list of forecast models. The forecast model is created based on a model formula in conjunction with other parameters, which is then used to forecast a specific number of periods ahead.

Usage

add_univariate_forecast_model(
  fc_models,
  ts_object_train,
  fc_name,
  fc_formula,
  model_fc = FALSE,
  periods_ahead = 1,
  periods_history = Inf,
  verbose = FALSE,
  log_message = ""
)

Arguments

fc_models	A named list of forecast models, with for each forecast model a list with the model itself and a table with forecast values.
ts_object_train	A time series object, which contains only the training data.
fc_name	A character string specifying the name to be used for the new model that is added to the list of existing forecast models.
fc_formula	A character string specifying the expression to be evaluated to train the time series forecast model.
model_fc	Boolean, which is to be set to TRUE if the forecast expression specified for fc_formula returns an object with a forecast method, or set to FALSE if the forecast expression returns a forecast directly.
periods_ahead	A positive integer value indicating the number of periods to forecast ahead.
periods_history	A positive integer value indicating the number of historic datapoints to use for training, which is only relevant for specific forecast methods such as drift and mean.
verbose	Boolean, which is set to TRUE if status updates are valued, or set to FALSE if they are not.

Value

A named list of forecast models, with for each forecast model a list with the model itself and a table with forecast values.

Examples

ts_object_train <- tstools::initialize_ts_forecast_data(
      data = dummy_gasprice,
      date_col = "year_month", 
      col_of_interest = "gasprice", 
      group_cols = c("state", "oil_company")
   ) %>% 
   dplyr::filter(grouping == "state = New York   &   oil_company = CompanyA") %>% 
   tstools::transform_data_to_ts_object()
add_univariate_forecast_model(
   fc_models = list(),
   ts_object_train = ts_object_train,
   fc_name = "fc_drift_l6m",
   fc_formula = "rwf(x, h, drift = TRUE)",
   model_fc = FALSE,
   periods_ahead = 12,
   periods_history = 6,
   verbose = T
)  

ing-bank/tsforecast documentation built on Sept. 18, 2020, 9:40 a.m.