Stochastic and genetic neural network combinations in trading and hybrid time-varying leverage effects

Sermpinis, G., Stasinakis, C. and Dunis, C. (2014) Stochastic and genetic neural network combinations in trading and hybrid time-varying leverage effects. Journal of International Financial Markets, Institutions and Money, 30(1), pp. 21-54. (doi:10.1016/j.intfin.2014.01.006)

[img]
Preview
Text
91501.pdf - Accepted Version

1MB

Abstract

The motivation of this paper is 3-fold. Firstly, we apply a Multi-Layer Perceptron (MLP), a Recurrent Neural Network (RNN) and aPsi-Sigma Network (PSN) architecture in a forecasting and trad-ing exercise on the EUR/USD, EUR/GBP and EUR/CHF exchangerates and explore the utility of Kalman Filter, Genetic Programming(GP) and Support Vector Regression (SVR) algorithms as forecastingcombination techniques. Secondly, we introduce a hybrid leveragefactor based on volatility forecasts and market shocks and studyif its application improves the trading performance of our mod-els. Thirdly, we introduce a specialized loss function for NeuralNetworks (NNs) in financial applications. In terms of our results,the PSN from the individual forecasts and the SVR from our forecastcombination techniques outperform their benchmarks in statisti-cal accuracy and trading efficiency. We also note that our tradingstrategy is successful, as it increased the trading performance ofmost of our models, while our NNs loss function seems promising.

Item Type:Articles
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Stasinakis, Dr Charalampos and Sermpinis, Professor Georgios
Authors: Sermpinis, G., Stasinakis, C., and Dunis, C.
College/School:College of Social Sciences > Adam Smith Business School > Economics
Journal Name:Journal of International Financial Markets, Institutions and Money
Publisher:Elsevier B.V.
ISSN:1042-4431
ISSN (Online):1873-0612
Copyright Holders:Copyright © 2014 Elsevier B.V.
First Published:First published in Journal of International Financial Markets, Institutions and Money 30(1):21-54
Publisher Policy:Reproduced in accordance with the copyright policy of the publisher.

University Staff: Request a correction | Enlighten Editors: Update this record