Beyond mathematical models and algorithms: quantitative investing explained in 8 key points
31 MAR, 2026
Bruno Taillardat, Head of Quantitative Investment at Edmond de Rothschild Asset Management
Equity markets present such a broad spectrum of information that a structured and clearly defined process is required to manage it. Quantitative portfolio construction allows this complexity to be organized and enables the analysis of large volumes of data in a consistent and transparent manner.
The quantitative approach reduces emotional biases and optimizes the use of investors’ long-term risk budgets. It helps channel managers’ intuition within a disciplined framework and optimally anticipate risks.
Experienced managers must challenge assumptions and incorporate forward-looking perspectives when necessary. It is essential to have a diverse team with extensive experience in quantitative strategies across different market contexts. Quantitative investing is most effective when systematic discipline and human oversight work together.
Although quantitative investing is often perceived as a mechanical process or too closely tied to limited indices and signals, the quantitative approach combines systematic construction with multiple sources of knowledge and intuition. It is based on a mindset of continuous research and innovation.
Machine learning techniques help define and identify the most relevant factors in changing contexts. Quantitative managers must continuously improve their processes to adapt to evolving markets and technologies.
Quantitative platforms enable testing new ideas and refining signals and portfolio construction techniques. Advances in data science and artificial intelligence make it possible to analyze larger and more multidimensional datasets.
Alternative data sources complement traditional financial information. These tools help identify patterns, understand risks, and adapt portfolios to increasingly complex and responsive markets. The development and implementation of more powerful technologies allow for the analysis of larger and multidimensional datasets, including alternative sources that complement traditional financial data. They facilitate pattern recognition, risk comprehension, and portfolio adaptation to more complex and reactive markets.
Human oversight is essential both in daily management and in the design and evolution of strategies. This oversight is especially critical in events that are difficult to model with historical data (regulatory shocks, geopolitical tensions, sanctions, changes in energy or climate policies). A clear example was the tariffs imposed on “Liberation Day” in the U.S. (April 2, 2025), which demonstrated how political decisions can rapidly alter sectoral or regional outlooks. In such cases, managers can adjust exposures, introduce temporary restrictions, and reevaluate assumptions and stress-test scenarios.
