Paper: arXiv 2412.09517

Abstract

Traditional methods employed in matrix volatility forecasting often overlook the inherent Riemannian manifold structure of symmetric positive definite matrices, treating them as elements of Euclidean space, which can lead to suboptimal predictive performance. Moreover, they often struggle to handle high-dimensional matrices. In this paper, we propose a novel approach for forecasting realized covariance matrices of asset returns using a Riemannian-geometry-aware deep learning framework. In this way, we account for the geometric properties of the covariance matrices, including possible non-linear dynamics and efficient handling of high-dimensionality. Moreover, building upon a Fréchet sample mean of realized covariance matrices, we are able to extend the HAR model to the matrix-variate. We demonstrate the efficacy of our approach using daily realized covariance matrices for the 50 most capitalized companies in the S&P 500 index, showing that our method outperforms traditional approaches in terms of predictive accuracy.

Complexity vs Empirical Score

  • Math Complexity: 9.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced Riemannian geometry, manifold theory, and custom geometric deep learning architectures (SPDNet), indicating very high mathematical density. It also validates the method on a substantial dataset (50 stocks from S&P 500) with specific predictive accuracy benchmarks and a portfolio optimization application, demonstrating strong empirical backing.

Research Flowchart

  flowchart TD
  Start["Research Goal: Forecast Realized Covariance Matrices<br>using Geometric Deep Learning"] --> Input["Data: Daily Realized Covariance Matrices<br>50 largest S&P 500 Equities"]
  Input --> Method["Methodology: Riemannian-Geometry-Aware<br>Deep Learning Framework"]
  Method --> Process["Process: Fréchet Sample Mean +<br>Matrix-Variate HAR Model Extension"]
  Process --> Manifold["Key Innovation: Accounts for Riemannian<br>Manifold Structure & High-Dimensionality"]
  Manifold --> Outcome["Outcome: Outperforms Traditional Methods<br>in Predictive Accuracy"]

Cited References (53)

  1. The distribution of realized exchange rate volatility (2001)
  2. Modeling and Forecasting Realized Volatility (2003)
  3. The distribution of exchange rate volatility (2001)
  4. A New Parametrization of Correlation Matrices (2021) — link
  5. Geometric Means in a Novel Vector Space Structure on Symmetric Positive‐Definite Matrices (2007) — link
  6. Forecasting co-volatilities via factor models with asymmetry and long memory in realized covariance (2015) — link
  7. Realized matrix-exponential stochastic volatility with asymmetry, long memory and higher-moment spillovers (2022) — link
  8. Fractionally integrated generalized autoregressive conditional heteroskedasticity (1996) — link
and 45 more
  1. Estimating Quadratic Variation Using Realised Variance (2002)
  2. Econometric Analysis of Realized Volatility and Its Use in Estimating Stochastic Volatility Models (2002)
  3. Forecasting multivariate realized stock market volatility (2011)
  4. Modeling and forecasting (un)reliable realized covariances for more reliable financial decisions (2018) — link
  5. Riemannian batch normalization for SPD neural networks (2019)
  6. Realized Volatility Forecasting with Neural Networks (2020)
  7. Comparing unconstrained parametrization methods for return covariance matrix prediction (2022)
  8. A statistical recurrent model on the manifold of symmetric positive definite matrices (2018)
  9. Modeling and Forecasting Large Realized Covariance Matrices and Portfolio Choice (2017) — link
  10. Modelling and Forecasting Multivariate Realized Volatility (2011)
  11. A Simple Approximate Long-Memory Model of Realized Volatility (2009)
  12. Optimal Versus Naive Diversification: How Inefficient is the 1/N Portfolio Strategy? (2009) — link
  13. Predicting the Daily Covariance Matrix for S$&$P 100 Stocks Using Intraday Data - But Which Frequency to Use? (2008)
  14. Non-Euclidean Statistics for Covariance Matrices, with Applications to Diffusion Tensor Imaging (2009) — link
  15. Procrustes analysis (2016) — link
  16. Dynamic conditional correlation: a simple class of multivariate GARCH models (2002)
  17. Multivariate simultaneous generalized ARCH (1995)
  18. Volatility is rough (2018) — link
  19. Generalized Procrustes analysis. (1975)
  20. Factor state–space models for high-dimensional realized covariance matrices of asset returns (2020) — link
  21. The Model Confidence Set (2011)
  22. A geometric framework for covariance dynamics (2022) — link
  23. From manifold to manifold: Geometry-aware dimensionality reduction for SPD matrices (2014)
  24. Computing a nearest symmetric positive semidefinite matrix (1988) — link
  25. A Riemannian network for SPD matrix learning (2017)
  26. Estimating Long Memory in Volatility (2005) — link
  27. Bayesian semiparametric modeling of realized covariance matrices (2016) — link
  28. A Multivariate Generalized Orthogonal Factor GARCH Model (2007) — link
  29. On loss functions and ranking forecasting performances of multivariate volatility models (2013)
  30. Riemannian-geometric regime-switching covariance hedging (2024) — link
  31. Electroencephalogram signals classification for sleep-state decision–a Riemannian geometry approach (2012)
  32. Portfolio selection (1952)
  33. Object Oriented Data Analysis (2022)
  34. Multivariate high-frequency-based volatility ( HEAVY ) models (2011)
  35. The Conditional Autoregressive F-Riesz Model for Realized Covariance Matrices (2024) — link
  36. Increasing the information content of realized volatility forecasts* (2023) — link
  37. Inference on the maximal rank of time-varying covariance matrices using high-frequency data (2021)
  38. Why does stock market volatility change over time? (1989)
  39. Pedestrian detection via classification on Riemannian manifolds (2008)
  40. GO-GARCH: A Multivariate Generalized Orthogonal GARCH Model (2002) — link
  41. A full-factor multivariate GARCH model (2003) — link
  42. Dreamnet: A deep Riemannian manifold network for spd matrix learning (2022)
  43. The Generalized Conditional Autoregressive Wishart Model for Multivariate Realized Volatility (2017) — link
  44. Graph-Based Methods for Forecasting Realized Covariances (2024) — link
  45. Machine Learning Classification of Riemannian Tangent Spaces Based on MI-BCI (2023)

0 of 53 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.