Cross-sectional momentum · Time-series trend · Regime filtersCompanies that have outperformed over the past 3-12 months tend to continue outperforming in the near future. This cross-sectional momentum pattern is one of the most robust anomalies in empirical finance. Our framework ranks the universe by risk-adjusted returns over multiple lookback windows and scores companies at the high and low ends of the distribution, with weighting reflecting signal strength and conviction.
Pure price momentum is vulnerable to sharp reversals — momentum crashes. Our data signals provide a fundamental overlay: we measure whether a company's momentum is corroborated by improving quality scores, rising information density, and positive sentiment trends. Momentum supported by improving fundamentals persists longer and reverses less violently than momentum driven purely by price dynamics.
Time-series trend following applies momentum logic to individual securities and broad market indices: signals are positive when price is above its moving average, neutral or negative when below. We combine this with regime detection from aggregate signals. When data across the universe signals a regime transition — from expansion to contraction — we adjust analytical parameters to account for the changed market dynamics.