Modeling exchange rates: Smooth transitions, neural networks, and linear models

Marcelo C. Medeiros; Álvaro Veiga; Carlos Eduardo Pedreira

doi:10.1109/72.935089

Modeling exchange rates: Smooth transitions, neural networks, and linear models

Marcelo C. Medeiros
, Álvaro Veiga
, Carlos Eduardo Pedreira

Research output: Contribution to journal › Article › peer-review

Abstract

The goal of this paper is to test for and model nonlinearities in several monthly exchange rates time series. We apply two different nonlinear alternatives, namely: the artificial neural-network time series model estimated with Bayesian regularization and a flexible smooth transition specification, called the neuro-coefficient smooth transition autoregression. The linearity test rejects the null hypothesis of linearity in 10 out of 14 series. We compare, using different measures, the forecasting performance of the nonlinear specifications with the linear autoregression and the random walk models.

Original language	English (US)
Pages (from-to)	755-764
Number of pages	10
Journal	IEEE Transactions on Neural Networks
Volume	12
Issue number	4
DOIs	https://doi.org/10.1109/72.935089
State	Published - Jul 2001
Externally published	Yes

Keywords

Bayesian regularization
Exchange rates
Neural networks
Smooth transition models
Time series

ASJC Scopus subject areas

Software
Computer Science Applications
Computer Networks and Communications
Artificial Intelligence

Online availability

10.1109/72.935089

Library availability

Discover UIUC Full Text

Cite this

@article{f97e870b3c4544f2b06fd59ff88cd5e2,

title = "Modeling exchange rates: Smooth transitions, neural networks, and linear models",

abstract = "The goal of this paper is to test for and model nonlinearities in several monthly exchange rates time series. We apply two different nonlinear alternatives, namely: the artificial neural-network time series model estimated with Bayesian regularization and a flexible smooth transition specification, called the neuro-coefficient smooth transition autoregression. The linearity test rejects the null hypothesis of linearity in 10 out of 14 series. We compare, using different measures, the forecasting performance of the nonlinear specifications with the linear autoregression and the random walk models.",

keywords = "Bayesian regularization, Exchange rates, Neural networks, Smooth transition models, Time series",

author = "Medeiros, \{Marcelo C.\} and {\'A}lvaro Veiga and Pedreira, \{Carlos Eduardo\}",

note = "The authors would like to thank T. Ter{\"a}svirta, R. Flores, and D. van Dijk for valuable comments, and the CNPq for the financial support. This work is part of the first author{\textquoteright}s Ph.D. dissertation at the Department of Electrical Engineering, Pontifical Catholic University of Rio de Janeiro, Brazil.",

year = "2001",

month = jul,

doi = "10.1109/72.935089",

language = "English (US)",

volume = "12",

pages = "755--764",

journal = "IEEE Transactions on Neural Networks",

issn = "1045-9227",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "4",

}

TY - JOUR

T1 - Modeling exchange rates

T2 - Smooth transitions, neural networks, and linear models

AU - Medeiros, Marcelo C.

AU - Veiga, Álvaro

AU - Pedreira, Carlos Eduardo

N1 - The authors would like to thank T. Teräsvirta, R. Flores, and D. van Dijk for valuable comments, and the CNPq for the financial support. This work is part of the first author’s Ph.D. dissertation at the Department of Electrical Engineering, Pontifical Catholic University of Rio de Janeiro, Brazil.

PY - 2001/7

Y1 - 2001/7

N2 - The goal of this paper is to test for and model nonlinearities in several monthly exchange rates time series. We apply two different nonlinear alternatives, namely: the artificial neural-network time series model estimated with Bayesian regularization and a flexible smooth transition specification, called the neuro-coefficient smooth transition autoregression. The linearity test rejects the null hypothesis of linearity in 10 out of 14 series. We compare, using different measures, the forecasting performance of the nonlinear specifications with the linear autoregression and the random walk models.

AB - The goal of this paper is to test for and model nonlinearities in several monthly exchange rates time series. We apply two different nonlinear alternatives, namely: the artificial neural-network time series model estimated with Bayesian regularization and a flexible smooth transition specification, called the neuro-coefficient smooth transition autoregression. The linearity test rejects the null hypothesis of linearity in 10 out of 14 series. We compare, using different measures, the forecasting performance of the nonlinear specifications with the linear autoregression and the random walk models.

KW - Bayesian regularization

KW - Exchange rates

KW - Neural networks

KW - Smooth transition models

KW - Time series

UR - https://www.scopus.com/pages/publications/0035392124

UR - https://www.scopus.com/pages/publications/0035392124#tab=citedBy

U2 - 10.1109/72.935089

DO - 10.1109/72.935089

M3 - Article

AN - SCOPUS:0035392124

SN - 1045-9227

VL - 12

SP - 755

EP - 764

JO - IEEE Transactions on Neural Networks

JF - IEEE Transactions on Neural Networks

IS - 4

ER -

Modeling exchange rates: Smooth transitions, neural networks, and linear models

Abstract

Keywords

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this