The RegressionTreeGB4TS class supports Gradient Boosting for Time Series data. Multi-horizon forecasting supported via the Recursive method. Given a response vector y, a predictor matrix x is built that consists of lagged y vectors. Additional future response vectors are built for training.
y_t = f(x)
where x = [y_{t-1}, y_{t-2}, ... y_{t-lags}].
Value parameters
- fname
-
the feature/variable names
- hparam
-
the hyper-parameters (use REGRESSION.hp for default)
- lags
-
the maximum lag included (inclusive)
- x
-
the input/predictor matrix built out of lags of y (and optionally from exogenous variables ex)
- yy
-
the output/response vector trimmed to match x.dim (@see ARX object)
Attributes
- Companion
- object
- Graph
-
- Supertypes
-
trait ForecasterXclass RegressionTreeGBtrait Fittrait FitMtrait Predictortrait FeatureSelectiontrait Modelclass Objecttrait Matchableclass AnyShow all
Members list
Value members
Concrete methods
Produce a vector of size h, of 1 through h-steps ahead forecasts for the model. forecast the following time points: t+1, ..., t-1+h. Note, must create the yf matrix before calling the forecast method. Intended to work with rolling validation (analog of predict method) Must call forecastAll first.
Produce a vector of size h, of 1 through h-steps ahead forecasts for the model. forecast the following time points: t+1, ..., t-1+h. Note, must create the yf matrix before calling the forecast method. Intended to work with rolling validation (analog of predict method) Must call forecastAll first.
Value parameters
- h
-
the forecasting horizon, number of steps ahead to produce forecasts
- t
-
the time point from which to make forecasts
- yf
-
the forecast matrix (time x horizons)
Attributes
Forecast values for all y_.dim time points at horizon h (h-steps ahead). Assign to FORECAST MATRIX and return h-step ahead forecast. Note, predictAll provides predictions for h = 1.
Forecast values for all y_.dim time points at horizon h (h-steps ahead). Assign to FORECAST MATRIX and return h-step ahead forecast. Note, predictAll provides predictions for h = 1.
Value parameters
- h
-
the forecasting horizon, number of steps ahead to produce forecasts
- yf
-
the forecast matrix for the endogenous variable y (time x horizons)
- yx
-
the matrix of endogenous y and exogenous x values
Attributes
- See also
-
forecastAllmethod inForecastertrait.
Get the internally row trimed and column expanded input matrix and reposnse vector.
Get the internally row trimed and column expanded input matrix and reposnse vector.
Attributes
Predict a value for y_t+1 using the 1-step ahead forecast. y_t+1 = f (y_t, ...) + e_t+1
Predict a value for y_t+1 using the 1-step ahead forecast. y_t+1 = f (y_t, ...) + e_t+1
Value parameters
- t
-
the time point from which to make prediction
- yx
-
the matrix of endogenous y and exogenous x values
Attributes
Test FORECASTS of a RegressionTreeGB4TS forecasting model y_ = f(x) + e and return its forecasts and QoF vector. Testing may be in-sample (on the training set) or out-of-sample (on the testing set) as determined by the parameters passed in. Note: must call train and forecastAll before testF.
Test FORECASTS of a RegressionTreeGB4TS forecasting model y_ = f(x) + e and return its forecasts and QoF vector. Testing may be in-sample (on the training set) or out-of-sample (on the testing set) as determined by the parameters passed in. Note: must call train and forecastAll before testF.
Value parameters
- h
-
the forecasting horizon, number of steps ahead to produce forecasts
- y_
-
the testing/full response/output vector
- yx
-
the matrix of endogenous y and exogenous x values
Attributes
Inherited methods
Perform backward elimination to find the least predictive variable to remove from the existing model, returning the variable to eliminate, the new parameter vector and the new Quality of Fit (QoF). May be called repeatedly.
Perform backward elimination to find the least predictive variable to remove from the existing model, returning the variable to eliminate, the new parameter vector and the new Quality of Fit (QoF). May be called repeatedly.
Value parameters
- cols
-
the columns of matrix x currently included in the existing model
- first
-
first variable to consider for elimination (default (1) assume intercept x_0 will be in any model)
- qk
-
index of Quality of Fit (QoF) to use for comparing quality
Attributes
- See also
-
Fitfor index of QoF measures. - Inherited from:
- Predictor
Perform backward elimination to find the least predictive variables to remove from the full model, returning the variables left and the new Quality of Fit (QoF) measures for all steps.
Perform backward elimination to find the least predictive variables to remove from the full model, returning the variables left and the new Quality of Fit (QoF) measures for all steps.
Value parameters
- cross
-
whether to include the cross-validation QoF measure
- first
-
first variable to consider for elimination
- qk
-
index of Quality of Fit (QoF) to use for comparing quality
Attributes
- See also
-
Fitfor index of QoF measures. - Inherited from:
- Predictor
Build a sub-model that is restricted to the given columns of the data matrix.
Build a sub-model that is restricted to the given columns of the data matrix.
Value parameters
- x_cols
-
the columns that the new model is restricted to
Attributes
- Definition Classes
- Inherited from:
- RegressionTreeGB
Attributes
- Inherited from:
- Predictor
Diagnose the health of the model by computing the Quality of Fit (QoF) measures, from the error/residual vector and the predicted & actual responses. For some models the instances may be weighted.
Diagnose the health of the model by computing the Quality of Fit (QoF) measures, from the error/residual vector and the predicted & actual responses. For some models the instances may be weighted.
Value parameters
- w
-
the weights on the instances (defaults to null)
- y_
-
the actual response/output vector to use (test/full)
- yp_
-
the predicted response/output vector (test/full)
Attributes
- See also
-
Regression_WLS - Definition Classes
- Inherited from:
- Fit
Diagnose the health of the model by computing the Quality of Fit (QoF) metrics/measures, from the error/residual vector and the predicted & actual responses. For some models the instances may be weighted. Include interval measures. Note: wis should be computed separately.
Diagnose the health of the model by computing the Quality of Fit (QoF) metrics/measures, from the error/residual vector and the predicted & actual responses. For some models the instances may be weighted. Include interval measures. Note: wis should be computed separately.
Value parameters
- alpha
-
the nominal level of uncertainty (alpha) (defaults to 0.9, 90%)
- low
-
the predicted lower bound
- up
-
the predicted upper bound
- w
-
the weights on the instances (defaults to null)
- y
-
the actual response/output vector to use (test/full)
- yp
-
the point prediction mean/median
Attributes
- See also
-
Regression_WLS - Inherited from:
- Fit
Diagnose the health of the model by computing the Quality of Fit (QoF) measures,
Diagnose the health of the model by computing the Quality of Fit (QoF) measures,
Value parameters
- alphas
-
the array of prediction levels
- low
-
the lower bounds for various alpha levels
- up
-
the upper bounds for various alpha levels
- y
-
the given time-series (must be aligned with the interval forecast)
- yp
-
the point prediction mean/median
Attributes
- Inherited from:
- Fit
Return the Quality of Fit (QoF) measures corresponding to the labels given. Note, if sse > sst, the model introduces errors and the rSq may be negative, otherwise, R^2 (rSq) ranges from 0 (weak) to 1 (strong). Override to add more quality of fit measures.
Return the Quality of Fit (QoF) measures corresponding to the labels given. Note, if sse > sst, the model introduces errors and the rSq may be negative, otherwise, R^2 (rSq) ranges from 0 (weak) to 1 (strong). Override to add more quality of fit measures.
Attributes
- Definition Classes
- Inherited from:
- Fit
Forecast values for all y_.dim time points and all horizons (1 through h-steps ahead). Record these in the FORECAST MATRIX yf, where yf(t, k) = k-steps ahead forecast for y_t Note, column 0, yf(?, 0), is set to y (the actual time-series values). last column, yf(?, h+1), is set to t (the time values, for reference). Forecast recursively down diagonals in the yf forecast matrix. The top right and bottom left triangles in yf matrix are not forecastable.
Forecast values for all y_.dim time points and all horizons (1 through h-steps ahead). Record these in the FORECAST MATRIX yf, where yf(t, k) = k-steps ahead forecast for y_t Note, column 0, yf(?, 0), is set to y (the actual time-series values). last column, yf(?, h+1), is set to t (the time values, for reference). Forecast recursively down diagonals in the yf forecast matrix. The top right and bottom left triangles in yf matrix are not forecastable.
Value parameters
- h
-
the maximum forecasting horizon, number of steps ahead to produce forecasts
- y_
-
the actual values to use in making forecasts
- yx
-
the matrix of endogenous y and exogenous x values
Attributes
- Inherited from:
- ForecasterX
Forecast intervals for all y_.dim time points at horizon h (h-steps ahead). Create prediction intervals (two vectors) for the given time points at level p. Caveat: assumes errors follow a Normal distribution. Override this method to handle other cases.
Forecast intervals for all y_.dim time points at horizon h (h-steps ahead). Create prediction intervals (two vectors) for the given time points at level p. Caveat: assumes errors follow a Normal distribution. Override this method to handle other cases.
Value parameters
- h
-
the forecasting horizon, number of steps ahead to produce forecasts
- p
-
the level (1 - alpha) for the prediction interval
- y_
-
the aligned actual values to use in making forecasts
- yfh
-
the forecast vector at horizon h
Attributes
- Inherited from:
- ForecasterX
Perform forward selection to find the most predictive variable to add the existing model, returning the variable to add and the new model. May be called repeatedly.
Perform forward selection to find the most predictive variable to add the existing model, returning the variable to add and the new model. May be called repeatedly.
Value parameters
- cols
-
the columns of matrix x currently included in the existing model
- qk
-
index of Quality of Fit (QoF) to use for comparing quality
Attributes
- See also
-
Fitfor index of QoF measures. - Inherited from:
- Predictor
Perform forward selection to find the most predictive variables to have in the model, returning the variables added and the new Quality of Fit (QoF) measures for all steps.
Perform forward selection to find the most predictive variables to have in the model, returning the variables added and the new Quality of Fit (QoF) measures for all steps.
Value parameters
- cross
-
whether to include the cross-validation QoF measure
- qk
-
index of Quality of Fit (QoF) to use for comparing quality
Attributes
- See also
-
Fitfor index of QoF measures. - Inherited from:
- Predictor
Run the full model before variable elimination as a starting point for backward elimination.
Run the full model before variable elimination as a starting point for backward elimination.
Value parameters
- qk
-
index of Quality of Fit (QoF) to use for comparing quality
Attributes
- Inherited from:
- Predictor
Return the best model found from feature selection.
Return the feature/variable names.
Return the used data matrix x. Mainly for derived classes where x is expanded from the given columns in x_, e.g., SymbolicRegression.quadratic adds squared columns.
Return the used data matrix x. Mainly for derived classes where x is expanded from the given columns in x_, e.g., SymbolicRegression.quadratic adds squared columns.
Attributes
- Inherited from:
- Predictor
Return the used response vector y. Mainly for derived classes where y is transformed, e.g., TranRegression, ARX.
Return the used response vector y. Mainly for derived classes where y is transformed, e.g., TranRegression, ARX.
Attributes
- Inherited from:
- Predictor
Return the the y-transformation.
Return the used response matrix y, if needed.
Return the used response matrix y, if needed.
Attributes
- See also
-
neuralnet.PredictorMV - Inherited from:
- Model
Return the help string that describes the Quality of Fit (QoF) measures provided by the Fit trait. Override to correspond to fitLabel.
Return the hyper-parameters.
Return the relative importance of selected variables, ordered highest to lowest, rescaled so the highest is one.
Return the relative importance of selected variables, ordered highest to lowest, rescaled so the highest is one.
Value parameters
- cols
-
the selected columns/features/variables
- rSq
-
the matrix R^2 values (stand in for sse)
Attributes
- Inherited from:
- Predictor
The log-likelihood function times -2. Override as needed.
The log-likelihood function times -2. Override as needed.
Value parameters
- ms
-
raw Mean Squared Error
- s2
-
MLE estimate of the population variance of the residuals
Attributes
- See also
- Inherited from:
- Fit
Return the mean of the squares for error (sse / df). Must call diagnose first.
Return the mean of the squares for error (sse / df). Must call diagnose first.
Attributes
- Inherited from:
- Fit
Return the number of terms/parameters in the model, e.g., b_0 + b_1 x_1 + b_2 x_2 has three terms.
Return the number of terms/parameters in the model, e.g., b_0 + b_1 x_1 + b_2 x_2 has three terms.
Attributes
- Inherited from:
- Predictor
Return the vector of parameter/coefficient values.
Given a data vector z, predict a value by summing the predicted values for each tree moderated by the learning rate eta.
Given a data vector z, predict a value by summing the predicted values for each tree moderated by the learning rate eta.
Value parameters
- z
-
the data vector to predict
Attributes
- Definition Classes
- Inherited from:
- RegressionTreeGB
Given a data matrix z, predict the value by summing the predict for each tree, for each row of the matrix.
Given a data matrix z, predict the value by summing the predict for each tree, for each row of the matrix.
Value parameters
- z
-
the data matrix to predict
Attributes
- Definition Classes
- Inherited from:
- RegressionTreeGB
Return the coefficient of determination (R^2). Must call diagnose first.
Return the coefficient of determination (R^2). Must call diagnose first.
Attributes
- Inherited from:
- FitM
Return a basic report on a trained and tested multi-variate model.
Return a basic report on a trained and tested multi-variate model.
Value parameters
- ftMat
-
the matrix of qof values produced by the
Fittrait
Attributes
- Inherited from:
- Model
Return a basic report on a trained and tested model.
Return a basic report on a trained and tested model.
Value parameters
- ftVec
-
the vector of qof values produced by the
Fittrait
Attributes
- Inherited from:
- Model
Reset the best-step to default
Reset the degrees of freedom to the new updated values. For some models, the degrees of freedom is not known until after the model is built.
Reset the degrees of freedom to the new updated values. For some models, the degrees of freedom is not known until after the model is built.
Value parameters
- df_update
-
the updated degrees of freedom (model, error)
Attributes
- Inherited from:
- Fit
Evalaute the model with only one column, e.g., intercept only model.
Evalaute the model with only one column, e.g., intercept only model.
Value parameters
- qk
-
index of Quality of Fit (QoF) to use for comparing quality
Attributes
- Inherited from:
- Predictor
Perform feature selection to find the most predictive features/variables to have in the model, returning the features/variables added and the new Quality of Fit (QoF) measures/metrics for all steps.
Perform feature selection to find the most predictive features/variables to have in the model, returning the features/variables added and the new Quality of Fit (QoF) measures/metrics for all steps.
Value parameters
- cross
-
whether to include the cross-validation QoF measure
- qk
-
index of Quality of Fit (QoF) to use for comparing quality
- tech
-
the feature selection technique to apply
Attributes
- See also
-
Fitfor index of QoF measures/metrics. - Inherited from:
- FeatureSelection
Show the prediction interval forecasts and relevant QoF metrics/measures.
Show the prediction interval forecasts and relevant QoF metrics/measures.
Value parameters
- h
-
the forecasting horizon
- low
-
the predicted lower bound
- qof_all
-
all the QoF metrics (for point and interval forecasts)
- up
-
the predicted upper bound
- yfh
-
the forecasts for horizon h
- yy
-
the aligned actual response/output vector to use (test/full)
Attributes
- Inherited from:
- Fit
Return the symmetric Mean Absolute Percentage Error (sMAPE) score. Caveat: y_i = yp_i = 0 => no error => no percentage error
Return the symmetric Mean Absolute Percentage Error (sMAPE) score. Caveat: y_i = yp_i = 0 => no error => no percentage error
Value parameters
- e_
-
the error/residual vector (if null, recompute)
- y
-
the given time-series (must be aligned with the forecast)
- yp
-
the forecasted time-series
Attributes
- Inherited from:
- FitM
Return the sum of the squares for error (sse). Must call diagnose first.
Return the sum of the squares for error (sse). Must call diagnose first.
Attributes
- Inherited from:
- FitM
Perform stepwise regression to find the most predictive variables to have in the model, returning the variables left and the new Quality of Fit (QoF) measures for all steps. At each step it calls forwardSel and backwardElim and takes the best of the two actions. Stops when neither action yields improvement.
Perform stepwise regression to find the most predictive variables to have in the model, returning the variables left and the new Quality of Fit (QoF) measures for all steps. At each step it calls forwardSel and backwardElim and takes the best of the two actions. Stops when neither action yields improvement.
Value parameters
- cross
-
whether to include the cross-validation QoF measure
- qk
-
index of Quality of Fit (QoF) to use for comparing quality
- swap
-
whether to allow a swap step (swap out a feature for a new feature in one step)
Attributes
- See also
-
Fitfor index of QoF measures. - Inherited from:
- Predictor
Produce a QoF summary for a model with diagnostics for each predictor x_j and the overall Quality of Fit (QoF). Note: `Fac_Cholesky is used to compute the inverse of xtx.
Produce a QoF summary for a model with diagnostics for each predictor x_j and the overall Quality of Fit (QoF). Note: `Fac_Cholesky is used to compute the inverse of xtx.
Value parameters
- b
-
the parameters/coefficients for the model
- fname
-
the array of feature/variable names
- vifs
-
the Variance Inflation Factors (VIFs)
- x_
-
the testing/full data/input matrix
Attributes
- Definition Classes
- Inherited from:
- Fit
Swap out variable with in variable.
Swap out variable with in variable.
Value parameters
- cols
-
the columns of matrix x currently included in the existing model
- in
-
the variable to swap in
- out
-
the variable to swap out
- qk
-
index of Quality of Fit (QoF) to use for comparing quality
Attributes
- Inherited from:
- Predictor
Test a predictive model y_ = f(x_) + e and return its QoF vector. Testing may be be in-sample (on the training set) or out-of-sample (on the testing set) as determined by the parameters passed in. Note: must call train before test.
Test a predictive model y_ = f(x_) + e and return its QoF vector. Testing may be be in-sample (on the training set) or out-of-sample (on the testing set) as determined by the parameters passed in. Note: must call train before test.
Value parameters
- x_
-
the testing/full data/input matrix (defaults to full x)
- y_
-
the testing/full response/output vector (defaults to full y)
Attributes
- Inherited from:
- RegressionTreeGB
Return the indices for the test-set.
Return the indices for the test-set.
Value parameters
- n_test
-
the size of test-set
- rando
-
whether to select indices randomly or in blocks
Attributes
- See also
-
scalation.mathstat.TnT_Split - Inherited from:
- Predictor
Use Gradient Boosting for Training. For every iteration, evaluate the residual and form a Regression Tree where the residual is the dependent value (equal to the gradient if using SSE as the loss function).
Use Gradient Boosting for Training. For every iteration, evaluate the residual and form a Regression Tree where the residual is the dependent value (equal to the gradient if using SSE as the loss function).
Value parameters
- x_
-
the training/full data/input matrix
- y_
-
the training/full response/output vector
Attributes
- Inherited from:
- RegressionTreeGB
The train2 method should work like the train method, but should also optimize hyper-parameters (e.g., shrinkage or learning rate). Only implementing classes needing this capability should override this method.
The train2 method should work like the train method, but should also optimize hyper-parameters (e.g., shrinkage or learning rate). Only implementing classes needing this capability should override this method.
Value parameters
- x_
-
the training/full data/input matrix (defaults to full x)
- y_
-
the training/full response/output vector (defaults to full y)
Attributes
- Inherited from:
- Predictor
Train and test the predictive model y_ = f(x_) + e and report its QoF and plot its predictions. Return the predictions and QoF. FIX - currently must override if y is transformed, @see TranRegression
Train and test the predictive model y_ = f(x_) + e and report its QoF and plot its predictions. Return the predictions and QoF. FIX - currently must override if y is transformed, @see TranRegression
Value parameters
- x_
-
the training/full data/input matrix (defaults to full x)
- xx
-
the testing/full data/input matrix (defaults to full x)
- y_
-
the training/full response/output vector (defaults to full y)
- yy
-
the testing/full response/output vector (defaults to full y)
Attributes
- Inherited from:
- Predictor
Attributes
- Inherited from:
- Predictor
Compute the Variance Inflation Factor (VIF) for each variable to test for multi-collinearity by regressing x_j against the rest of the variables. A VIF over 50 indicates that over 98% of the variance of x_j can be predicted from the other variables, so x_j may be a candidate for removal from the model. Note: override this method to use a superior regression technique.
Compute the Variance Inflation Factor (VIF) for each variable to test for multi-collinearity by regressing x_j against the rest of the variables. A VIF over 50 indicates that over 98% of the variance of x_j can be predicted from the other variables, so x_j may be a candidate for removal from the model. Note: override this method to use a superior regression technique.
Value parameters
- skip
-
the number of columns of x at the beginning to skip in computing VIF
Attributes
- Inherited from:
- Predictor
Inherited fields
The optional reference to an ontological concept