Fundamentals

Micro alphas are small, often short-lived predictive signals that carry little edge on their own and only become useful in combination. This pillar sets the shared vocabulary and first-principles methods the rest of the site assumes: what a micro alpha actually is, how signal research is structured, how a candidate signal is validated against noise, and how its contribution is attributed once it is live.

The articles are meant to be read as a progression. Start with what a micro alpha is and where systematic strategies look for one, then move to how signal discovery is organized as a research process. From there, statistical validation covers how to decide whether a weak signal is real or an artifact of the search, and attribution closes the loop by separating a signal's genuine contribution from overlap with factors you already hold.

Read together, these pieces are the toolkit you reach for before touching portfolio construction, execution, or machine-learning methods. They are deliberately method-first and stack-agnostic: the goal is to make your downstream choices defensible, not to prescribe a single workflow.