| predict.srlTS {srlTS} | R Documentation |
Predict function for srlTS object
Description
Predict function for srlTS object
Usage
## S3 method for class 'srlTS'
predict(object, n_ahead = 1, X_test, y_test, cumulative = 0, ...)
Arguments
object |
an srlTS object |
n_ahead |
number of times ahead to predict by iteration |
X_test |
a matrix exogenous features |
y_test |
the test series; needed for future predictions (optional; see details) |
cumulative |
should cumulative (rolling) sums be returned (integer indicating number of times to sum) |
... |
currently unused |
Details
The 'y_test' argument must be supplied if forecasts are desired or if 'n_ahead' < 'nrow(X_test)'. This is because in order to obtain 1-step forecast for, say, the 10th observation in the test data set, the 9th observation of 'y_test' is required. The length of 'y_test' will determine how many forecasts to produce. In order to get true forecasts for the first 30 observations after the training set, one must (currently) produce the set of 1-step, 2-step, 3-step, ..., 30-step ahead predictions.
Value
a vector of predictions
Examples
data("LakeHuron")
fit_LH <- srlTS(LakeHuron)
predict(fit_LH)