temp_holdout: Create Temporal Crossvalidated Dataset
In pmontman/fforma: Feature-based Forecast Method Averaging

Description Usage Arguments Details Value

View source: R/ensemble_classifier.R

Creates a datasets of time series for forecasting and metalearning by extracting the final observations of each series

1	temp_holdout(dataset)

dataset

A list with each element having at least the following

x: A time series object ts with the historical data.
h: The number of required forecasts.

The final h observation are extrated an posed as the true future values in the entry xx of the output list. At least 7 observations are kept as the observable time series x.

A list with the same structure as the input, the following entries may be added or overwritten if existing

x: A time series object ts with the historical data.
xx: A time series with the true future data. Has length h unless the remaining x would be too short.