Dynamic Ensembles for Time Series Forecasting
Sequential Re-weighting for controlling predictions' redundancy
Sliding similarity via Pearson's correlation
Soft Imputation
Computing the softmax
Splitting expressions by pattern
Training procedure of for ADE
ADE training poor version Train meta-models in the training data, as o...
Dynamic Ensembles for Time Series Forecasting
Arbitrated Dynamic Ensemble
Arbitrated Dynamic Ensemble
Predictions by an ADE ensemble
Predictions by an ADE ensemble
Computing the absolute error
base_ensemble-class
base_ensemble
Computing the error of base models
Get best PLS/PCR model
Prequential Procedure in Blocks
Fit Cubist models (M5)
Fit Feedforward Neural Networks models
Fit Gaussian Process models
Fit Generalized Boosted Regression models
Fit Generalized Linear Models
Fit Multivariate Adaptive Regression Splines models
Fit PLS/PCR regression models
Fit Projection Pursuit Regression models
Fit Random Forest models
Fit Support Vector Regression models
Base model for XGBoost
Wrapper for creating an ensemble
Building a committee for an ADE model
Combining the predictions of several models
Compute the predictions of base models
Dynamic Ensemble for Time Series
Dynamic Ensemble for Time Series
Predictions by an DETS ensemble
Predictions by an DETS ensemble
Weighting Base Models by their Moving Average Squared Error
Embedding a Time Series
Get the target from a formula
Extract top learners from their weights
Get the response values from a data matrix
Holdout
Out-of-bag loss estimations
Out-of-bag predictions
Applying lapply on the rows
Training the base models of an ensemble
Training an arbiter
Training a RBR arbiter
Arbiter predictions via Cubist
Training a Gaussian prosadacess arbiter
Arbiter predictions via linear ssmodel
Training a Gaussian process arbiter
Arbiter predictions via linear model
Training a LASSO arbiter
Arbiter predictions via linear model
Training a meta_mars process arbiter
Arbiter predictions via mars model
Training a pls process arbiter
Arbiter predictions via pls model
Training a meta_mars process arbiter
Arbiter predictions via ppr model
Predicting loss using arbiter
Training a random forest arbiter
Arbiter predictions via ranger
Training a Gaussian process arbiter
Arbiter predictions via linear model
Training a xgb arbiter
Arbiter predictions via xgb
Recent performance of models using EMASE
Setup base learning models
Setup base learning models
Model weighting
Computing the mean squared error
Scale a numeric vector using max-min
Predicting new observations using an ensemble
predict method for pls/pcr
Computing the proportions of a numeric vector
rbind with do.call syntax
Get most recent lambda observations
Computing the root mean squared error
Computing the rolling mean of the columns of a matrix
Computing the squared error
Selecting best model according to weights
Updating an ADE model
Updating the metalearning layer of an ADE model
Update the base models of an ensemble
Updating the weights of base models
XGB optimizer
XGBoost predict function
asdasd
A framework for dynamically combining forecasting models for time series forecasting predictive tasks. It leverages machine learning models from other packages to automatically combine expert advice using metalearning and other state-of-the-art forecasting combination approaches. The predictive methods receive a data matrix as input, representing an embedded time series, and return a predictive ensemble model. The ensemble use generic functions 'predict()' and 'forecast()' to forecast future values of the time series. Moreover, an ensemble can be updated using methods, such as 'update_weights()' or 'update_base_models()'. A complete description of the methods can be found in: Cerqueira, V., Torgo, L., Pinto, F., and Soares, C. "Arbitrated Ensemble for Time Series Forecasting." to appear at: Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Springer International Publishing, 2017; and Cerqueira, V., Torgo, L., and Soares, C.: "Arbitrated Ensemble for Solar Radiation Forecasting." International Work-Conference on Artificial Neural Networks. Springer, 2017 <doi:10.1007/978-3-319-59153-7_62>.