Using Conditional GANs to Tune and Combine Trading Strategies

Summary

This paper proposes conditional generative adversarial networks for calibrating individual trading strategies and combining their signals. It lays out a workflow for training and selecting a cGAN for time-series data, using generated samples to tune strategies, and using the collection of samples in ensemble modeling. The aim is to improve strategy settings or aggregate weaker signals into a portfolio-level approach.

The authors evaluate the method with multiple trading strategies across 579 assets, comparing it with an ensemble approach and time-series validation methods. They report that cGANs can be a suitable alternative and that they outperform when traditional techniques fail to produce alpha. The summary does not identify the assets, evaluation period, trading costs, or robustness checks, so it offers limited detail for judging generalization or implementation. Its evidence is experimental and should be interpreted within the tested setup.

Key ideas

• Conditional GANs are proposed for strategy calibration and signal aggregation.
• The workflow covers cGAN training and selection for time-series data.
• Generated samples are used both to tune strategies and to build an ensemble.
• The experiment spans multiple strategies and 579 assets, with time-series comparisons.
• The reported results suggest an advantage when traditional methods do not find alpha.

Tags

# Generative Adversarial Networks for Financial Trading Strategies Fine-Tuning and Combination


# Generative Adversarial Networks for Financial Trading Strategies Fine-Tuning and Combination









Systematic trading strategies are algorithmic procedures that allocate assets aiming to optimize a certain performance criterion. To obtain an edge in a highly competitive environment, the analyst needs to proper fine-tune its strategy, or discover how to combine weak signals in novel alpha creating manners. Both aspects, namely fine-tuning and combination, have been extensively researched using several methods, but emerging techniques such as Generative Adversarial Networks can have an impact into such aspects. Therefore, our work proposes the use of Conditional Generative Adversarial Networks (cGANs) for trading strategies calibration and aggregation. To this purpose, we provide a full methodology on: (i) the training and selection of a cGAN for time series data; (ii) how each sample is used for strategies calibration; and (iii) how all generated samples can be used for ensemble modelling. To provide evidence that our approach is well grounded, we have designed an experiment with multiple trading strategies, encompassing 579 assets. We compared cGAN with an ensemble scheme and model validation methods, both suited for time series. Our results suggest that cGANs are a suitable alternative for strategies calibration and combination, providing outperformance when the traditional techniques fail to generate any alpha.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.