Hi! Here's Iván from Noax Capital 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 two must-read academic papers that mix cutting-edge ML/DL with Asset Pricing & Quant Finance:

  • Predicting Weekly Stock Market Movements with Machine Learning

  • Harnessing Generative AI for Economic Insights

• 💊 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 introduce a novel approach to capture the low-volatility anomaly more efficiently through a leveraged low-risk strategy. The paper demonstrates how this strategy outperforms both market benchmarks and traditional low-volatility portfolios.

Today’s Sponsor

Savvy Investors Know Where to Get Their News—Do You?

Here’s the truth: there is no magic formula when it comes to building wealth.

Much of the mainstream financial media is designed to drive traffic, not good decision-making. Whether it’s disingenuous headlines or relentless scare tactics used to generate clicks, modern business news was not built to serve individual investors.

Luckily, we have The Daily Upside. Created by Wall Street insiders and bankers, this fresh, insightful newsletter delivers valuable insights that go beyond the headlines.

And the best part? It’s completely free. Join 1M+ readers and subscribe today.

“Predicting Weekly Stock Market Movements with Machine Learning”

🔔 This paper introduces a new approach to predicting weekly stock market movements using machine learning techniques, with a focus on medium-term trends and reduced computational requirements.

The study outperforms traditional benchmarks by introducing new features and a unique random trader benchmark, demonstrating the potential of ML in weekly market predictions.

Key takeaways 👇

➡ The paper focuses on weekly predictions rather than daily, offering a broader perspective on medium-term trends. A new benchmark of random traders is introduced, providing a more objective standard for performance evaluation.

➡ Novel features are incorporated, including scaling laws and directional changes, alongside traditional technical indicators.

➡ The training process involves adjusting datasets by assigning varying weights to different samples. Multiple models are trained and tested, with the multi-layer perceptron (MLP) showing stability across various market trends.

➡ The multi-layer perceptron (MLP) model demonstrates stability and robustness across various market trends, including upward, downward, and cyclic movements.

“Harnessing Generative AI for Economic Insights”

📢 AI-Powered Managerial Expectations for Economic Forecasting. A novel approach to predicting economic trends? Keep reading! 🔽 

This paper introduces an innovative method using generative AI to extract managerial expectations from corporate conference call transcripts, creating a powerful tool for economic forecasting.

The study's AI Economy Score significantly outperforms traditional benchmarks and survey forecasts, demonstrating the potential of AI-extracted managerial insights for both macro and microeconomic predictions.

Key takeaways:

➡ The analysis includes over 120,000 conference call transcripts from 5,513 unique companies, providing a vast dataset for extracting economic insights.

➡ The AI Economy Score robustly predicts future GDP growth, adding 4% to the R-squared of models that include common predictors like Term Spread and Real FFR.

➡ This predictive power persists for up to 4 quarters, indicating that the AI Economy Score can forecast GDP growth effectively over the short term.

➡ Industry-level measures derived from the AI Economy Score show significant predictive power for GDP growth lasting up to 4 years, aiding long-term economic decision-making.

➡ Positive shocks to the AI Economy Score predict higher consumption, investment, and output growth for at least 8 quarters, along with increased inflation and market returns.

➡ Firm-level expectation scores derived from this method can predict company sales and earnings for up to 4 years, showcasing versatility across different economic scales.

➡ The AI-extracted managerial expectations provide a cost-effective alternative to traditional surveys, with the potential to complement existing forecasting methods by offering unique insights into future economic activities.

This innovative approach leverages generative AI to provide valuable forward-looking information that can enhance economic research and policy-making.

Today’s Sponsor

Learn how to make AI work for you

AI won’t take your job, but a person using AI might. That’s why 800,000+ professionals read The Rundown AI – the free newsletter that keeps you updated on the latest AI news and teaches you how to use it in just 5 minutes a day.

Sign up to start learning.

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

“Low-Risk Alpha Without Low Beta”

🔔📈 The paper proposes a novel approach to capture the low-volatility anomaly more efficiently. Keep reading! 👇

The paper introduces a leveraged low-risk strategy that outperforms both the market and traditional low-volatility portfolios.

Key takeaways:

➡️ The strategy leverages multi-factor low-risk portfolios to a beta of 1.0 while controlling tracking error.

➡️ In developed markets (1986-2023), the strategy's information ratio (IR) increased from 0.43 to 0.92.

➡️ Outperformance jumped from 3.32% to 5.92% relative to the benchmark.

➡️ Ex-post tracking error decreased from 7.75% to 6.43%.

➡️ Sharpe ratio slightly decreased from 0.72 to 0.67, but maximum benchmark-relative drawdown dropped from 22% to 12%.

➡️ Outperformance stems from the low-risk tilt, not leverage effects.

➡️ The approach is robust across geographies and market environments.

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! 🌟🤖💼