HRP: Building a Portfolio with Machine Learning

Tired of portfolios that collapse at the slightest market shock? 📉

In this video, I present to you the core of my investment strategy: Hierarchical Risk Parity (HRP).

Developed by Marcos López de Prado, this system uses Machine Learning to build a robust portfolio. Unlike classic methods, HRP does not rest on uncertain return forecasts, but on the real risk structure of your assets.

On the agenda:

• The concept: Why the correlation between companies (such as NVIDIA, IONQ, or Caterpillar) is more important than you think.
• The method: How the system groups assets by risk “families” to avoid concentration.
• Analysis of my list: Why I chose names such as Bloom Energy, Volvo, and PTC in this specific framework.
• Putting it into practice: Why this system is more stable than a classic 60/40 allocation.

The goal? Less volatility, more peace of mind, and mathematically proven diversification. 🧠

My tools:

• 🐍 Python & yfinance for data retrieval.
• 📊 Optimization via hierarchical clustering.