Evaluating data augmentation for financial time series classification

Elizabeth Fons¹

Paula Dawson²

Xiao-jun Zeng¹

John Keane¹

Alexandros Iosifidis³

• Paper
• Code

Abstract

Data augmentation methods in combination with deep neural networks have been used extensively in computer vision on classification tasks, achieving great success; however, their use in time series classification is still at an early stage. This is even more so in the field of financial prediction, where data tends to be small, noisy and non-stationary. In this paper we evaluate several augmentation methods applied to stocks datasets using two state-of-the-art deep learning models. The results show that several augmentation methods significantly improve financial performance when used in combination with a trading strategy. For a relatively small dataset ($\approx 30K$$\approx30K$ samples), augmentation methods achieve up to $400\mathrm{\%}$$400\%$ improvement in risk adjusted return performance; for a larger stock dataset ($\approx 300K$$\approx300K$ samples), results show up to $40\mathrm{\%}$$40\%$ improvement.

Paper

Bibtex

x
@misc{fons2020-da-finance,
title={Evaluating data augmentation for financial time series classification}, 
author={Elizabeth Fons and Paula Dawson and Xiao-jun Zeng and John Keane and Alexandros           Iosifidis},
year={2020},
eprint={2010.15111},
archivePrefix={arXiv},
primaryClass={q-fin.ST}
}

Acknowledgement

This work was supported by the European Union's Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie Grant Agreement no. 675044 (http://bigdatafinance.eu/), Training for Big Data in Financial Research and Risk Management. A. Iosifidis acknowledges funding from the Independent Research Fund Denmark project DISPA (Project Number: 9041-00004).