RMT-calibrated financial eigenvector dynamics: separating genuine covariance rotation from sampling noise, testing forecastability on Grassmann and flag manifolds, and building geometry-aware covariance forecasts.
quantitative-finance manifold-learning spectral-analysis random-matrix-theory time-series-forecasting riemannian-geometry financial-machine-learning flag-manifold covariance-forecasting eigenvector-dynamics matrix-perturbation-theory
-
Updated
Aug 5, 2026 - Python