The Predictor companion object provides a method for testing predictive models.
Attributes
Members list
Value members
Concrete methods
Make the PREDICTION INTERVALS for the given model.
Make the PREDICTION INTERVALS for the given model.
Value parameters
- mod
-
the model making the predictions
- xx
-
the aligned data/input matrix to use (test/full)
- yp
-
the corresponding vector of predicted values
- yy
-
the aligned actual response/output vector to use (test/full)
Attributes
Plot the actual and predicted values/vectors both ordered by increasing yy values.
Plot the actual and predicted values/vectors both ordered by increasing yy values.
Value parameters
- doPlot
-
whether to plot y-actual vs. predictions
- mName
-
the model name
- order
-
whether to order all vectors by y-actual
- yp
-
the corresponding vector of predicted values
- yy
-
the aligned actual response/output vector to use (test/full)
Attributes
Plot the PREDICTION INTERVALS with all vectors ordered by increasing yp values.
Plot the PREDICTION INTERVALS with all vectors ordered by increasing yp values.
Value parameters
- doPlot
-
whether to plot y-actual vs. predictions as well as prediction intervals
- low_up
-
the predicted (lower, upper) bound vectors
- mName
-
the model name
- order
-
whether to order all vectors by y-actual
- yp
-
the corresponding vector of predicted values
- yy
-
the aligned actual response/output vector to use (test/full)
Attributes
Test (in-sample) by training and testing on the FULL dataset. Test (out-of-sample) by training on the TRAINING set and testing on the TESTING set.
Test (in-sample) by training and testing on the FULL dataset. Test (out-of-sample) by training on the TRAINING set and testing on the TESTING set.
Value parameters
- check
-
whether to check the assertion that the in-sample and out-of-sample results are in rough agreement (e.g., at 20%)
- ext
-
the model subtype extension (e.g., indicating the transformation function used)
- mod
-
the model to be used