An Ensemble of LSTM Neural Networks for High-Frequency Stock Market Classification

27 Pages Posted: 16 Jul 2018 Last revised: 15 Feb 2019

See all articles by Svetlana Borovkova

Svetlana Borovkova

Vrije Universiteit Amsterdam - Faculty of Economics and Business Administration

Ioannis Tsiamas

VU University Amsterdam

Date Written: June 25, 2018

Abstract

We propose an ensemble of Long-Short Term Memory (LSTM) Neural Networks for intraday stock predictions, using a large variety of Technical Analysis indicators as network inputs. The proposed ensemble operates in an online way, weighting the individual models proportionally to their recent performance, which allows us to deal with possible non-stationarities in an innovative way. The performance of the models is measured by Area Under the Curve of the Receiver Operating Characteristic. We evaluate the predictive power of our model on several US large-cap stocks and benchmark it against Lasso and Ridge logistic classifiers. The proposed model is found to perform better than the benchmark models or equally weighted ensembles.

Keywords: High-Frequency Trading, Deep Learning, LSTM Neural Networks, Ensemble Models

JEL Classification: C45, C53, C55, G17

Suggested Citation

Borovkova, Svetlana and Tsiamas, Ioannis, An Ensemble of LSTM Neural Networks for High-Frequency Stock Market Classification (June 25, 2018). Available at SSRN: https://ssrn.com/abstract=3202313 or http://dx.doi.org/10.2139/ssrn.3202313

Svetlana Borovkova (Contact Author)

Vrije Universiteit Amsterdam - Faculty of Economics and Business Administration ( email )

De Boelelaan 1105
Amsterdam, 1081HV
Netherlands

Ioannis Tsiamas

VU University Amsterdam ( email )

De Boelelaan 1105
Amsterdam, ND North Holland 1081 HV
Netherlands

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
622
Abstract Views
1,957
rank
46,311
PlumX Metrics