Spatio-Temporal Momentum

Hi! Here's Iván with this week's exciting newsletter, brimming with insights and discoveries on building robust investment strategies and risk models using Machine Learning.

In this edition, I am presenting the following sections:

• 🕹️ AI-Finance Insights: I summarize three must-read academic papers that mix cutting-edge ML/DL with Asset Pricing & Quant Finance:

  • Spatio-Temporal Momentum

  • DeepScalper: A Reinforcement Learning Framework to Capture Fleeting Intraday Trading Opportunities

• 💊 AI Essentials: The section on top AI & Quant Finance learning resources: Today, I'm thrilled to present a great video with hands-on guidance and complete code for using Deep Learning Momentum with TFT for trading. This must-watch video offers over an hour step-by-step instructions, helping you understand and apply these advanced techniques in your trading strategies.

• 🥐 Asset Pricing Insights: In this edition, I recommend the study "Machine Learning vs. Economic Restrictions: Evidence from Stock Return Predictability," which examines the effectiveness of machine learning models compared to traditional economic restrictions.

Today’s Sponsor

The Rising Demand for Whiskey: A Smart Investor’s Choice

Why are 250,000 Vinovest customers investing in whiskey?

In a word - consumption.

Global alcohol consumption is on the rise, with projections hitting new peaks by 2028. Whiskey, in particular, is experiencing significant growth, with the number of US craft distilleries quadrupling in the past decade. Younger generations are moving from beer to cocktails, boosting whiskey's popularity.

That’s not all.

Whiskey's tangible nature, market resilience, and Vinovest’s strategic approach make whiskey a smart addition to any diversified portfolio.

Schedule a call with a Vinovest advisor to learn more.

“Spatio-Temporal Momentum”

The paper introduces a novel approach that merges time-series and cross-sectional momentum strategies through the innovative use of neural networks.

The main contributions and findings are as follows:

👉 The study presents Spatio-Temporal Momentum (STMOM) strategies, integrating both time-series and cross-sectional momentum features to create a unified trading signal for each asset. This approach leverages the relationship between temporal and cross-sectional momentum, providing a holistic view.

👉 The researchers demonstrate that a simple neural network with a single fully connected layer effectively learns to generate trading signals for all portfolio assets by incorporating their momentum features. This simplicity outperforms more complex models in terms of both interpretability and performance.

👉 Backtesting on portfolios of 46 actively-traded US equities and 12 equity index futures contracts shows that the STMOM model maintains superior performance compared to traditional benchmarks, even under high transaction costs of up to 5-10 basis points.

👉 Coupling the STMOM model with least absolute shrinkage and turnover regularization yields the best performance across various transaction cost scenarios, highlighting the robustness and efficiency of this approach.

👉 Traditional momentum strategies like Time-Series Momentum (TSMOM) and Cross-Sectional Momentum (CSMOM) are less effective compared to the STMOM approach. The Single Layer Perceptron (SLP) model, in particular, shows significant improvements in profitability and risk-adjusted performance metrics.

“DeepScalper: A Reinforcement Learning Framework to Capture Fleeting Intraday Trading Opportunities”

The paper introduces DeepScalper, a cutting-edge intraday trading framework leveraging deep reinforcement learning (RL) to optimize trading decisions in high-dimensional action spaces.

The main contributions and findings are as follows:

👉 Advanced Optimization: DeepScalper uses a dueling Q-network with action branching and an encoder-decoder architecture to process macro and micro-level market data, optimizing trading strategies efficiently.

👉 Risk-Aware Trading: The framework includes a unique reward function with a hindsight bonus and a risk-aware auxiliary task, balancing profit and risk effectively to enhance long-term decision-making.

👉 Superior Performance: Empirical results demonstrate that DeepScalper outperforms state-of-the-art baselines in key financial metrics, such as total return, Sharpe ratio, Calmar ratio, and Sortino ratio, using real-world data from six financial futures over three years.

👉 Comprehensive Validation: Extensive experiments and ablation studies confirm the significant contributions of each component, validating DeepScalper's superior performance and practical applicability in intraday trading.

Sponsor 👇

Million dollar AI strategies packed in this free 3 hour AI Masterclass – designed for founders & professionals. Act fast because It’s free only for the first 100.

Join it here for $0. 🎁

AI-Essentials

Great video with hands-on guidance and complete code for using Deep Learning Momentum with TFT for trading. Over an hour of step-by-step instructions. 👇

Asset Pricing Insights

“Machine Learning vs. Economic Restrictions: Evidence from Stock Return Predictability”

Can machine learning models outperform traditional methods in stock return prediction?

This paper by Avramov, Cheng, and Metzker (published in Management Science) examines the effectiveness of machine learning models compared to traditional economic restrictions.

The study uses techniques like neural networks, feed-forward networks, recurrent neural networks with LSTM cells, and generative adversarial networks to predict returns for a broad sample of U.S. stocks from 1987 to 2017.

The findings show that machine learning models can achieve high monthly returns without restrictions, with the NN3 model delivering an average value-weighted long-short portfolio return of 1.56% per month and an FF6-adjusted return of 0.92%.

However, when economic restrictions like excluding microcaps and distressed firms are applied, performance significantly drops. For example, the value-weighted FF6-adjusted return for the NN3 model decreases by 66% after excluding microcaps.

Additionally, high turnover rates exceeding 87% per month suggest substantial trading costs, affecting the strategies' economic viability.

If you're enjoying our newsletter and want to support us, please recommend it to anyone you know who's interested in AI and Finance. Your referrals are the biggest compliment and help us grow! 🌟🤖💼