Improving forecasting accuracy of time series data using a new ARIMA-ANN hybrid method and empirical mode decomposition.
Ümit Çavus BüyüksahinSeyda ErtekinPublished in: CoRR (2018)
Keyphrases
- hybrid method
- empirical mode decomposition
- artificial neural networks
- hybrid model
- forecasting accuracy
- absolute percentage error
- non stationary
- neural network model
- bp neural network
- demand forecasting
- neural network
- back propagation
- support vector machine
- hybrid algorithm
- computational intelligence
- wavelet decomposition
- forecasting model
- hilbert huang transform
- genetic algorithm
- radial basis function
- support vector regression
- multi band
- intrinsic mode functions
- genetic algorithm ga
- support vector machine svm
- multiscale
- image processing
- short term
- training speed
- feature extraction