Skip to content
All library documents

Using Deep Learning to Predict Market Trends Beyond Historical Momentum

Article arXiv papers · Author: Fernando Berzal et al.
Read the original

Summary

The paper argues that conventional trend-following and momentum strategies rely on past price behavior and may be limited when market conditions change. It proposes using artificial intelligence and machine-learning techniques to predict future market trends, with the aim of informing asset-management decisions.

The authors suggest that accurate forecasts could help increase returns and reduce drawdowns, but the supplied text does not describe a particular model, input data, evaluation design, or measured results. The claimed benefits should therefore be read as motivation rather than demonstrated evidence in this excerpt. It gives a high-level rationale for predictive trend models, but not enough detail to judge their robustness, implementation, or performance against established strategies.

Key ideas

  • The paper presents historical trend-following and momentum as limited by their reliance on past market behavior.
  • It advocates machine-learning methods to forecast future market trends.
  • The proposed forecasts are intended to support asset-management decisions.
  • The excerpt provides no model details or empirical evidence for the suggested return and drawdown benefits.

Tags

Machine learningTrend followingMomentumRisk management
Full text
# Beyond Trend Following: Deep Learning for Market Trend Prediction


# Beyond Trend Following: Deep Learning for Market Trend Prediction









Trend following and momentum investing are common strategies employed by asset managers. Even though they can be helpful in the proper situations, they are limited in the sense that they work just by looking at past, as if we were driving with our focus on the rearview mirror. In this paper, we advocate for the use of Artificial Intelligence and Machine Learning techniques to predict future market trends. These predictions, when done properly, can improve the performance of asset managers by increasing returns and reducing drawdowns.

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.