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

Carlos Alberto Franco Franco, Juan Carlos Figueroa-Garcia, eduyn ramiro Lopéz-Santana

Research output: Contribution to journal › 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.
Original languageEnglish (US)
Pages (from-to)786-795
Number of pages10
JournalLecture Notes in Computer Science
Volume10361
StatePublished - 2017

Fingerprint

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

Cite this

Franco Franco, Carlos Alberto ; Figueroa-Garcia, Juan Carlos ; Lopéz-Santana, eduyn ramiro. / A Three-Step Deep Neural Network Methodology for Exchange Rate Forecasting. In: Lecture Notes in Computer Science. 2017 ; Vol. 10361. pp. 786-795.
@article{856ea2eb75cb498d8d3da30b8638dff6,
title = "A Three-Step Deep Neural Network Methodology for Exchange Rate Forecasting",
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.",
author = "{Franco Franco}, {Carlos Alberto} and Figueroa-Garcia, {Juan Carlos} and Lop{\'e}z-Santana, {eduyn ramiro}",
year = "2017",
language = "English (US)",
volume = "10361",
pages = "786--795",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer",

}

Franco Franco, CA, Figueroa-Garcia, JC & Lopéz-Santana, ER 2017, 'A Three-Step Deep Neural Network Methodology for Exchange Rate Forecasting', Lecture Notes in Computer Science, vol. 10361, pp. 786-795.

A Three-Step Deep Neural Network Methodology for Exchange Rate Forecasting. / Franco Franco, Carlos Alberto; Figueroa-Garcia, Juan Carlos ; Lopéz-Santana, eduyn ramiro.

In: Lecture Notes in Computer Science, Vol. 10361, 2017, p. 786-795.

Research output: Contribution to journal › Article

TY - JOUR

T1 - A Three-Step Deep Neural Network Methodology for Exchange Rate Forecasting

AU - Franco Franco, Carlos Alberto

AU - Figueroa-Garcia, Juan Carlos

AU - Lopéz-Santana, eduyn ramiro

PY - 2017

Y1 - 2017

N2 - 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.

AB - 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.

M3 - Article

VL - 10361

SP - 786

EP - 795

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -

Franco Franco CA, Figueroa-Garcia JC, Lopéz-Santana ER. A Three-Step Deep Neural Network Methodology for Exchange Rate Forecasting. Lecture Notes in Computer Science. 2017;10361:786-795.