Enhancing Time-Series Momentum Strategies Using Deep Neural Networks
By Bryan Lim, Stefan Zohren, Stephen J. Roberts. 2019
103 citations recorded by the source corpus.
Although time-series momentum is a well-studied phenomenon in finance, common strategies require the explicit definition of both a trend estimator and a position sizing rule. In this article, the authors introduce deep momentum networks—a hybrid approach that injects deep learning–based trading rules into the volatility scaling framework of time-series momentum. The model also simultaneously learns both trend estimation and position sizing in a data-driven manner, with networks directly trained by optimizing the Sharpe ratio of the signal. Backtesting on a portfolio of 88 continuous futures contracts, the authors demonstrate that the Sharpe-optimized long short-term memory improved traditional methods by more than two times in the absence of transactions costs and continued outperforming when considering transaction costs up to 2–3 bps. To account for more illiquid assets, the authors also propose a turnover regularization term that trains the network to factor in costs at run-time. TOPICS: Statistical methods, simulations, big data/machine learning Key Findings • While time-series momentum strategies have been extensively studied in f inance, common strategies require the explicit specification of a trend estimator and position sizing rule. • In this article, the authors introduce deep momentum networks —a hybrid approach that injects deep learning–based trading rules into the volatility scaling framework of timeseries momentum. • Backtesting on a portfolio of continuous futures contracts, Deep Momentum Networks were shown to outperform traditional methods for transaction costs of up to 2–3 bps, with a turnover regularisation term proposed for more illiquid assets.
Subfields: Time-Series & Forecasting
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Time-Series & Forecasting
Citations in the corpus
- Time-series forecasting with deep learning: a survey (2020, methodology): “In general, hybrid models use deep neural networks in two manners: a) to encode time-varying parameters for non-probabilistic parametric models [64,65,67], and b) to produce distribution parameters for probabilistic models [38,40,66].”