AI and algorithmic trading
Machine learning applied to markets promises a lot and often fails, for specific reasons: little signal, plenty of noise, and data whose properties shift over time.
These articles separate the uses that hold up from those that are mostly marketing.
AI Algorithmic Trading: Real Uses vs. Marketing Hype
Between marketing promises and real-world use, where does AI in algorithmic trading actually stand today? A hype-free assessment of concrete contributions and structural limits.
Overfitting in Machine Learning Trading: Why So Many Models Fail
A model that predicts the past perfectly and collapses in live conditions: overfitting is the number-one trap in machine learning applied to trading. Understanding why, and how to guard against it.
Robo-Advisors: How Automated Allocation Actually Works
Behind a robo-advisor's simplified interface lies an allocation mechanism based on well-established quantitative models. Here's what actually happens between your risk profile questionnaire and your portfolio's final allocation.