The ForecasterX companion object provides functions useful for classes extending the ForecasterX trait, i.e., forecasting models with multiple input variables.
Attributes
- Companion
- trait
- Graph
-
- Supertypes
-
class Objecttrait Matchableclass Any
- Self type
-
ForecasterX.type
Members list
Value members
Concrete methods
Evaluate the quality of point and optionally interval forecast for horizon (h = 1 to hh).
Evaluate the quality of point and optionally interval forecast for horizon (h = 1 to hh).
Value parameters
- hh
-
the maximum forecasting horizon (h = 1 to hh)
- ints
-
whether to evaluate prediction interval forecasts as well as point forecasts
- mod
-
the forecasting model to be evaluated
- y
-
the testing/full response/output vector
- yx
-
the matrix of endogenous y and exogenous x values
Attributes
Use rolling-validation to compute test Quality of Fit (QoF) measures by dividing the dataset into a TRAINING SET (tr) and a TESTING SET (te) as follows: [ <-- tr_size --> | <-- te_size --> ] This version calls predict (RECURSIVE) for h-steps ahead out-of-sample forecasts.
Use rolling-validation to compute test Quality of Fit (QoF) measures by dividing the dataset into a TRAINING SET (tr) and a TESTING SET (te) as follows: [ <-- tr_size --> | <-- te_size --> ] This version calls predict (RECURSIVE) for h-steps ahead out-of-sample forecasts.
Value parameters
- hh
-
the max forecasting horizon (h = 1, 2, ... hh)
- mod
-
the forecasting model being used (e.g.,
RegressionTreeRF4TS) - rc
-
the retraining cycle (number of forecasts until retraining occurs)
- te_size
-
the size of the testing set (negative => use ratio to calculate
Attributes
- See also
-
RollingValidation
Split the x matrix and y vector into training and testing sets.
Split the x matrix and y vector into training and testing sets.
Value parameters
- ratio
-
the ratio of the TESTING set to the full dataset (most common 80-20, 70-30)
- x
-
the x data/input matrix
- y
-
the y response/output vector