• Corpus ID: 2004015

AN IMPROVED MULTILAYER PERCEPTRON BASED ON WAVELET APPROACH FOR PHYSICAL TIME SERIES PREDICTION

@inproceedings{Ali2014ANIM,
  title={AN IMPROVED MULTILAYER PERCEPTRON BASED ON WAVELET APPROACH FOR PHYSICAL TIME SERIES PREDICTION},
  author={Ashikin Ali},
  year={2014}
}
The real world datasets engage many challenges such as noisy data, periodic variations on several scales and long-term trends that do not vary periodically. Meanwhile, Neural Networks (NN) has been successfully applied in many problems in the domain of time series prediction. The standard NN adopts computationally intensive training algorithms and can easily get trapped into local minima. To overcome such drawbacks in ordinary NN, this study focuses on using a wavelet technique as a… 

Figures from this paper

References

SHOWING 1-10 OF 144 REFERENCES

The wavelet multilayer perceptron for the prediction of earthquake time series data

This work proposed a new model which makes use the wavelet technique to pre-process the time series data before feeding to the MLP, and it is called a wavelet multilayer perceptron (W-MLP).

Combining Neural Network Forecasts on Wavelet-transformed Time Series

We discuss a simple strategy aimed at improving neural network prediction accuracy, based on the combination of predictions at varying resolution levels of the domain under investigation (here: time

A novel hybridization of artificial neural networks and ARIMA models for time series forecasting

Wavelet-based Multiresolution Forecasting

Results of modelling and forecasting nonstationary financial time series using a combination of the maximal overlap discreet wavelet transform (MODWT) and fuzzy logic and fuzzy-rule bases are discussed.

The wavelet filtering in temperature time series prediction

Simulation results on the prediction of temperature time series show that W-MLP performs considerably better results when compared to other four filtering techniques in terms of the prediction error and epochs.

Time series forecasting using a hybrid ARIMA and neural network model

  • G. Zhang
  • Computer Science
    Neurocomputing
  • 2003

FINANCIAL TIME SERIES FORECASTING BY NEURAL NETWORK USING CONJUGATE GRADIENT LEARNING ALGORITHM AND MULTIPLE LINEAR REGRESSION WEIGHT INITIALIZATION

The proposed conjugate gradient with MLR weight initialization requires a lower computation cost and learns better than backpropagation with random initialization and the results find that neural networks can model the time series satisfactorily.

Multiresolution forecasting for futures trading using wavelet decompositions

This work investigates the effectiveness of a financial time-series forecasting strategy which exploits the multiresolution property of the wavelet transform to choose short past windows for the inputs to the MLPs at lower scales and long past windows at higher scales.
...