Prediction of hierarchical time series using structured regularization and its application to artificial neural networks.
Tomokaze ShiratoriKen KobayashiYuichi TakanoPublished in: CoRR (2020)
Keyphrases
- artificial neural networks
- prediction accuracy
- financial time series
- evolutionary artificial neural networks
- ann models
- predictive model
- neural network
- moving average
- computational intelligence
- chaotic time series
- neural network model
- exponential smoothing
- feed forward artificial neural networks
- prediction model
- dynamic time warping
- non stationary
- max margin learning
- multi layer perceptron
- general loss functions
- structured data
- back propagation
- short term prediction
- elman network
- using artificial neural networks
- stock market
- forecasting accuracy
- regularization term
- neural models
- extreme values
- prediction algorithm
- hybrid model
- hidden layer
- prediction error
- recurrent neural networks
- multilayer perceptron
- short term
- real world