arima forecastor — arima_forecastor • kassandr

Envelope function that forecasts arima model

arima_forecastor(
  fit,
  test_sample = NA,
  predicted = "",
  predicted_ = "",
  options = "",
  predictors = "",
  h = 1,
  frequency = 1
)

Arguments

fit	fitted arima model
test_sample	tsibble with predictors for test sample
predicted	name of all the predicted variables. Not used as is determined by the fit.
predicted_	name of the perdicted variable. Not used as is determined by the fit.
options	character with options separated by comma not supparted
predictors	character with predictors separated by plus or comma, no predictors by default
h	forecasting horizon
frequency	time series frequency may be it is needed for forecasting with regressors, may check later

Value

forecast object

Details

Envelope function that forecasts arima model.

Examples

train_sample = tsibble::tsibble(date = lubridate::ymd("2017-01-01") + months(0:99),
    y = arima.sim(n = 100, model = list(ar = 0.7)), x = 1:100)
#> Using `date` as index variable.
fit = arima_estimator(train_sample, "y", "p=2,d=0,q=0", "x")
#> Error in stats::arima(x = x, order = order, seasonal = seasonal, xreg = xreg,  : 
#>   lengths of 'x' and 'xreg' do not match
test_sample = tsibble::tsibble(date = lubridate::ymd("2030-01-01") + months(0:1), x = 5:6)
#> Using `date` as index variable.
arima_forecastor(fit, test_sample, predictors = "x", h = 2, frequency = 12)
#> [1] NA