ATAforecasting (version 0.0.60)

ATA.Forecast: Forecasting Method for The ATAforecasting

Description

ATA.Forecast is a generic function for forecasting of the ATA Method.

Usage

ATA.Forecast(
  object,
  h = NULL,
  out.sample = NULL,
  ci.level = 95,
  negative.forecast = TRUE,
  onestep = FALSE,
  print.out = TRUE
)

Value

An object of class ata and forecast values.

Arguments

object

An ata object is required for forecast.

h

Number of periods for forecasting.

out.sample

A numeric vector or time series of class ts or msts for out-sample.

ci.level

Confidence Interval levels for forecasting. Default value is 95.

negative.forecast

Negative values are allowed for forecasting. Default value is TRUE. If FALSE, all negative values for forecasting are set to 0.

onestep

Default is FALSE. if TRUE, the dynamic forecast strategy uses a one-step model multiple times (h forecast horizon) where the prediction for the prior time step is used as an input for making a prediction on the following time step.

print.out

Default is TRUE. If FALSE, forecast summary of ATA Method is not shown.

Author

Ali Sabri Taylan and Hanife Taylan Selamlar

References

#'yapar2017msesATAforecasting

#'yapar2018mhesATAforecasting

#'yapar2018msesATAforecasting

#'yapar2019ataATAforecasting

See Also

forecast, stlplus, stR, stl, decompose, tbats, seasadj.

Examples

trainATA <-  head(touristTR, 84)
ata_fit <- ATA(trainATA, parPHI = 1, seasonal.test = TRUE, seasonal.model = "decomp")
ata_fc <- ATA.Forecast(ata_fit, h=12)