Sermpinis, Georgios and Verousis, Thanos and Theofilatos, Konstantinos (2016) 'Adaptive Evolutionary Neural Networks for Forecasting and Trading without a Data-Snooping Bias.' Journal of Forecasting, 35 (1). 1 - 12. ISSN 0277-6693

Preview

Text Adaptive Evolutionary Neural Networks for Forecasting and Trading without a Data Snooping Bias Verousis.pdf - Accepted Version Download (615kB) | Preview

Abstract

In this paper, we present two neural‐network‐based techniques: an adaptive evolutionary multilayer perceptron (aDEMLP) and an adaptive evolutionary wavelet neural network (aDEWNN). The two models are applied to the task of forecasting and trading the SPDR Dow Jones Industrial Average (DIA), the iShares NYSE Composite Index Fund (NYC) and the SPDR S&P 500 (SPY) exchange‐traded funds (ETFs). We benchmark their performance against two traditional MLP and WNN architectures, a smooth transition autoregressive model (STAR), a moving average convergence/divergence model (MACD) and a random walk model. We show that the proposed architectures present superior forecasting and trading performance compared to the benchmarks and are free from the limitations of the traditional neural networks such as the data‐snooping bias and the time‐consuming and biased processes involved in optimizing their parameters

Item Type:	Article
Subjects:	H Social Sciences > HG Finance
Divisions:	Faculty of Social Sciences > Essex Business School Faculty of Social Sciences > Essex Business School > Essex Finance Centre
Depositing User:	Elements
Date Deposited:	15 Mar 2019 15:21
Last Modified:	15 Mar 2019 16:15
URI:	http://repository.essex.ac.uk/id/eprint/24180

Actions (login required)

View Item