A Three-Step Deep Neural Network Methodology for Exchange Rate Forecasting Academic Article

abstract

  • We present a methodology for volatile time series forecasting using deep learning. We use a three-step methodology in order to remove trend and nonlinearities from data before applying two parallel deep neural networks to forecast two main features from processed data: absolute value and sign. The proposal is successfully applied to a volatile exchange rate time series problem.

publication date

  • 2017/1/1

edition

  • 10361

keywords

  • Absolute value
  • Deep learning
  • Deep neural networks
  • Exchange rate
  • Forecast
  • Forecasting
  • Learning
  • Methodology
  • Neural Networks
  • Nonlinearity
  • Time Series Forecasting
  • Time series
  • Trends
  • Volatiles

International Standard Serial Number (ISSN)

  • 0302-9743

number of pages

  • 10

start page

  • 786

end page

  • 795