Paper: arXiv 2603.17463

Authors: Massimiliano Caporin, Daniele Girolimetto, Emanuele Lopetuso

Abstract

We assess the advantage of combining univariate and multivariate portfolio risk forecasts with the aid of forecast reconciliation techniques. In our analyzes, we assume knowledge of portfolio weights, a standard for portfolio risk management applications. With an extensive simulation experiment, we show that, if the true covariance is known, forecast reconciliation improves over a standard multivariate approach, in particular when the adopted multivariate model is misspecified. However, if noisy proxies are used, correctly specified models and the misspecified ones (for instance, neglecting spillovers) turn out to be, in several cases, indistinguishable, with forecast reconciliation still providing improvements. The noise in the covariance proxy plays a crucial role in determining the improvement of both the forecast reconciliation and the correct model specification. An empirical analysis shows how forecast reconciliation can be adopted with real data to improve traditional GARCH-based portfolio variance forecasts.

Complexity vs Empirical Score

  • Math Complexity: 6.5/10
  • Empirical Rigor: 7.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: This paper effectively bridges theoretical econometric modeling and practical risk management by applying forecast reconciliation to multivariate GARCH models. The extensive simulation design and real-world empirical validation demonstrate strong methodological rigor and clear practical utility.

Research Flowchart

  flowchart TD
    A["Research Goal: Improve Portfolio Variance Forecasts via Forecast Reconciliation"] --> B["Data & Inputs: Portfolio Weights, Univariate & Multivariate GARCH Models"]
    B --> C["Simulation & Empirical Analyses"]
    C --> D["Computational Process: Reconcile Univariate & Multivariate Risk Forecasts"]
    D --> E["Evaluations: True Covariance vs. Noisy Proxies vs. Empirical Data"]
    E --> F["Outcome: Reconciliation beats standard multivariate models, especially under misspecification"]
    E --> G["Outcome: Proxy noise strongly impacts model distinction and reconciliation gains"]

Cited References (88)

  1. Improving cross-temporal forecasts reconciliation accuracy and utility in energy market (2025) — link
  2. Modeling financial contagion using mutually exciting jump processes (2015) — link
  3. High-Frequency Financial Econometrics (2014)
  4. Principal Component Models for Generating Large GARCH Covariance Matrices (2002) — link
  5. The Distribution of Realized Stock Return Volatility (2001) — link
  6. The Distribution of Realized Exchange Rate Volatility (2001) — link
  7. Modeling and Forecasting Realized Volatility (2003) — link
  8. Roughing It Up: Including Jump Components in the Measurement, Modeling, and Forecasting of Return Volatility (2007) — link
and 80 more
  1. No-Arbitrage Semi-Martingale Restrictions for Continuous-Time Volatility Models Subject to Leverage Effects, Jumps and i.i.d. Noise: Theory and Testable Distributional Implications (2007) — link
  2. A reduced form framework for modeling volatility of speculative prices based on realized variation measures (2011) — link
  3. Heteroskedasticity and Autocorrelation Consistent Covariance Matrix Estimation (1991) — link
  4. Forecasting with temporal hierarchies (2017) — link
  5. Forecast reconciliation: A review (2024) — link
  6. Measuring downside risk: Realised semivariance (2010)
  7. Multivariate GARCH models: A survey (2005) — link
  8. Networks in risk spillovers: A multivariate GARCH perspective (2023) — link
  9. Modelling the Coherence in Short-Run Nominal Exchange Rates: A Multivariate Generalized ARCH Model (1990) — link
  10. Realized Semi(co)variation: Signs That All Volatilities are Not Created Equal (2022) — link
  11. Exploiting the errors: A simple approach for improved volatility forecasting (2016) — link
  12. From zero to hero: Realized partial (co)variances (2022) — link
  13. Multiple comparisons among means (1961) — link
  14. Robust Estimation of Intraweek Periodicity in Volatility and Jump Detection (2011) — link
  15. The Estimation of Large Social Account Matrices (1978) — link
  16. Corrigenda: The Estimation of Large Social Account Matrices (1979) — link
  17. On the linkages between stock prices and exchange rates: Evidence from the banking crisis of 2007–2010 (2014) — link
  18. Scalar BEKK and Indirect DCC (2008) — link
  19. The Role of Jumps in Realized Volatility Modeling and Forecasting (2023) — link
  20. Exploiting Intraday Decompositions in Realized Volatility Forecasting: A Forecast Reconciliation Approach (2024) — link
  21. Proximity-structured multivariate volatility models (2015) — link
  22. A Practical Guide to Harnessing the HAR Volatility Model (2021) — link
  23. Forecast combination puzzle in the HAR model (2021) — link
  24. Forecast combination puzzle in the HAR model (2023) — link
  25. Negative volatility spillovers in the unrestricted ECCC-GARCH model (2010) — link
  26. A Simple Approximate Long-Memory Model of Realized Volatility (2009) — link
  27. The Volatility of Realized Volatility (2008) — link
  28. Discrete-Time Volatility Forecasting With Persistent Leverage Effect and the Link With Continuous-Time Volatility Modeling (2012) — link
  29. Measuring forecasting accuracy: The case of judgmental adjustments to SKU-level demand forecasts (2013) — link
  30. Comparing predictive accuracy (1995) — link
  31. Measuring financial asset return and volatility spillovers, with application to global equity markets (2009) — link
  32. On the past, present, and future of the Diebold-Yilmaz approach to dynamic network connectedness (2023) — link
  33. Forecast Combination-Based Forecast Reconciliation: Insights and Extensions (2024) — link
  34. Cross-Temporal Forecast Reconciliation: Optimal Combination Method and Heuristic Alternatives (2023) — link
  35. Large scale conditional covariance modeling, estimation and testing matrix (2001)
  36. Autoregressive Conditional Heteroskedasticity with estimates of the variance of United Kingdom inflation (1982) — link
  37. Dynamic Conditional Correlation: A Simple Class of Multivariate Generalized Autoregressive Conditional Heteroskedasticity Models (2002) — link
  38. Multivariate simultaneous generalized ARCH (1995) — link
  39. FoReco : Forecast reconciliation (2026) — link
  40. EUROMIND: a monthly indicator of the euro area economic conditions (2011) — link
  41. GARCH Models: Structure, Statistical Inference and Financial Applications (2019)
  42. Point and probabilistic forecast reconciliation for general linearly constrained multiple time series (2024) — link
  43. Cross-temporal probabilistic forecast reconciliation: Methodological and practical issues (2023) — link
  44. EuroMInd-C: a Disaggregate Monthly Indicator of Economic Activity for the Euro Area and member countries (2015) — link
  45. A Test for Superior Predictive Ability (2005) — link
  46. The Model Confidence Set (2011) — link
  47. An extended constant conditional correlation GARCH model and its fourth-moment structure (2004) — link
  48. Optimal combination forecasts for hierarchical time series (2011) — link
  49. Fast computation of reconciled forecasts for hierarchical and grouped time series (2016) — link
  50. The M3 Competition: Statistical Tests of the Results (2005) — link
  51. Cross-Temporal Coherent Forecasts for Australian Tourism (2019) — link
  52. On loss function and ranking forecasting performances of multivariate volatility models (2013) — link
  53. A well conditioned estimator for large-dimensional covariance matrices (2004) — link
  54. A forecast reconciliation approach to cause-of-death mortality modeling (2019) — link
  55. Asymptotic theory for a vector ARMA-GARCH model (2003) — link
  56. Matrix differential calculus with applications in statistics and econometrics. (2019)
  57. M5 Accuracy Competition: Results , Findings, and Conclusions (2022) — link
  58. Macroeconomic forecasting in the Euro area: country specific versus area-wide information (2003) — link
  59. Improving out-of-sample forecasts of stock price indexes with forecast reconciliation and clustering (2025) — link
  60. Forecasting hierarchical time series in supply chains: An empirical investigation (2022) — link
  61. Forecast reconciliation: A geometric view with new insights on bias correction (2021) — link
  62. Probabilistic Forecast Reconciliation: Properties, Evaluation and Score Optimisation (2023) — link
  63. Volatility forecast comparison using imperfect volatility proxies (2011) — link
  64. Data-based ranking of realised volatility estimators (2011) — link
  65. Optimal combination of realised volatility estimators (2009) — link
  66. Good Volatility, Bad Volatility: Signed Jumps and the Persistence of Volatility (2015) — link
  67. `Horses for Courses’ in demand forecasting (2014) — link
  68. The stationary bootstrap (1994) — link
  69. The effects of disaggregation on forecasting nonstationary time series (2014) — link
  70. Rsolnp : General Non-linear Optimization Using Augmented Lagrange Multiplier Method (2015) — link
  71. A shrinkage approach to large-scale covariance matrix estimation and implications for functional genomics (2005) — link
  72. Forecasting the volatility of crude oil futures using intraday data (2014) — link
  73. Grouped Functional Time Series Forecasting: An Application to Age-Specific Mortality Rates (2017) — link
  74. A bottom-up methodology for long term electricity consumption forecasting of an industrial sector - Application to pulp and paper sector in Brazil (2018) — link
  75. Multivariate GARCH models (2009)
  76. Hierarchical forecasting based on AR-GARCH model in a coherent structure (2007) — link
  77. The Precision of National Income Estimates (1942) — link
  78. Game- Theoretically Optimal Reconciliation of Contemporaneous Hierarchical Time Series Forecasts (2015) — link
  79. A bottom-up short-term residential load forecasting approach based on appliance characteristic analysis and multi-task learning (2021) — link
  80. Optimal Forecast Reconciliation for Hierarchical and Grouped Time Series Through Trace Minimization (2019) — link

0 of 88 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.