error_vin: Forecast Error of the Direct Forecast From Vintage
In nelson16silva/wavdrcast: Wavelet-Based Direct Multistep Forecast

Description Usage Arguments Details Value See Also Examples

Compute the root-mean-squared forecast error (RMSE) or mean absolute forecast error (MAE) of the direct forecast from vintage. This dataset is defined as the sequence of values that represent the latest estimate of the data at a particular moment in time.

error_vin(
  .vin,
  lags,
  .H,
  .K,
  ic = BIC,
  .var = "all",
  .diff = FALSE,
  dev = "RMSE",
  wav = FALSE,
  xreg = NULL,
  ...
)

`.vin`	List of data vintages (data frame) for each variable. The first vintage is related to the dependent variable and the number of columns of each data frame must be the same as the number of the out-of-sample forecasts (`.K`). The latest observation must be the last column of the data frame. If `wav = TRUE`, the length of the list can be one, otherwise, must be greater than one.
`lags`	A integer vector defining the lags of the regressors. If `wav = FALSE`, the length of the vector `lags` have to be the same as the length of `.vin`. However, if `wav = TRUE`, an additional element in `lags` must be add such that the last element in the vector `lags` is the lag of the wavelet regressor.
`.H`	An integer representing the maximum horizon step.
`.K`	An integer, the number of pseudo-out-of-sample forecasts.
`ic`	Information criterion, `BIC` or `AIC`. When searching for the best model the dataset is adjusted so that every model have the same data length for appropriate comparasion.
`.var`	A string to determine how the model will be specificated: `"all"` (default), `"ar"` or `"ar_out"`.
`.diff`	Logical `FALSE` or `TRUE`. If `TRUE`, the dependent variable is differentiated. See `wavdrcast-package` for the implication on the model specification and wavelet variable.
`dev`	A string, `"RMSE"` or `"MAE"`.
`wav`	Logical. If `TRUE`, a wavelet-based signal is add to `df`, where the signal is related to the explained variable, the first column of `df`.
`xreg`	Data frame. Exogeunous variable not subjected to be lagged. The number of rows must be the same as in `df`.
`...`	Further arguments passed to `wavsigmap::signal`.

See wavdrcast-package and model.

A double vector with forecasts error (RMSE or MAE) from 1 to .H step-ahead.

model, fcast, error

set.seed(1)

v1 <- tibble::tibble(v1_t1 = c(sample(100, 98, replace = TRUE), NA, NA),
                     v1_t2 = c(v1_t1[1:98], 500, NA),
                     v1_t3 = c(v1_t2[1:99], 280))

 set.seed(1)
v2 <- tibble::tibble(v2_t1 = 0.4 * v1[["v1_t1"]] + rnorm(100),
                     v2_t2 = c(v2_t1[1:98], 500 + rnorm(1), NA),
                     v2_t3 = c(v2_t2[1:99], 600 + rnorm(1)))

error_vin(list(v1, v2), lags = c(2, 2), .H = 3, .K = 3)
error_vin(list(v1), lags = c(2, 2), .H = 3, .K = 3, wav = TRUE)