Sector rotation matrix
A heatmap of sector-level returns across multiple lookback windows (typically 1M / 3M / 12M for the 11 GICS sectors), used to characterize whether leadership is rotating early-cycle, late-cycle, or defensive.
Cell[sector, window] = sector total return − benchmark total return over `window`The sector rotation matrix is a visualization aid, not a forecast. It plots relative-return spreads for the 11 GICS sectors against the broad market across a short, medium, and long lookback window — typically one, three, and twelve months. Each cell is tinted bull or bear by the sign of the spread; absolute magnitudes below a configurable neutral threshold render as flat to suppress noise. The reason this matters is that the historical pattern of sector leadership maps loosely to the business cycle: early-cycle regimes reward financials, consumer discretionary, and industrials; late-cycle regimes reward energy, materials, and staples; contractions reward utilities, staples, and healthcare. The matrix is the cleanest way to ask which of those patterns the market is actually pricing right now. The model has no predictive component — sector momentum is a notoriously poor cross-sectional return predictor — but it provides a common reference frame for the regime read and the scenario tree. Use it to corroborate the regime band, not to override it.