resamplr: Resampling methods

Description Usage Arguments Details Value Methods (by class) References

Generate test/train set for time series. Each training set consists of observations before the test set. This is also called "evaluation on a rolling forecasting origin".

crossv_ts(data, ...)

## S3 method for class 'data.frame'
crossv_ts(data, horizon = 1L, test_size = 1L,
  test_partial = FALSE, train_partial = TRUE, train_size = n,
  test_start = NULL, from = 1L, to = n, by = 1L, ...)

## S3 method for class 'grouped_df'
crossv_ts(data, horizon = 1L, test_size = 1L,
  test_partial = FALSE, train_partial = TRUE, train_size = n,
  test_start = NULL, from = 1L, to = n, by = 1L, ...)

`data`	A data frame
`...`	Arguments passed to methods
`horizon`	Difference between the first test set observation and the last training set observation
`test_size`	Size of the test set
`test_partial`	If `TRUE`, then allows for partial test sets. If `FALSE`, does not allow for partial test sets. If a number, then it is the minimum allowable size of a test set.
`train_partial`	Same as `test_partial`, but for the training set.
`train_size`	The maximum size of the training set. This allows for a rolling window training set, instead of using all obsservations from the start of the time series.
`test_start, from, to, by`	An integer vector of the starting index values of the test set. `NULL`, then these are generated from `seq(from, to, by)`.

In time-series cross-validation the training set only uses observations that are prior to the test set. Suppose the time series has n observations, the training set has a maximum size of r <= n and minimum size of s >= r. and the test set has a maximum size of p <= n and minimum size of q >= p. For indices i \in \{1, …, N\}:

Select observations i, …, \max{p, n} for the test set.
Select observations \max{i - h - p}, …, i - h for the training set.
If the test set has a size of at least q and the training set has a size of at least r.

A data frame with k the following columns:

sample: A list of resample objects. Training sets.
.id: An integer vector of identifiers.

data.frame: Data frame method. This rows are assumed to be ordered.
grouped_df: Grouped data frame method. The groups are assumed to be ordered, and the cross validation works on groups rather than rows.

Hyndman RJ (2017). forecast: Forecasting functions for time series and linear models. R package version 8.0, URL.
Hyndman RJ and Khandakar Y (2008). "Automatic time series forecasting: the forecast package for R." Journal of Statistical Software. URL.
Rob J. Hyndman. http://robjhyndman.com/hyndsight/tscv/. December 5, 2016.
Rob J. Hyndman. http://robjhyndman.com/hyndsight/tscvexample/. August 26, 2011.
Rob J. Hyndman and George Athanasopoulos. "Evaluating Forecast Accuracy." URL.
Max Kuhn. "Data splitting for Time Series." The caret Package. 2016-11-29. URL.