Paper: arXiv 2605.12099

Authors: Patrick Woitschig, Mike West

Abstract

We present a new class of Bayesian dynamic models for bivariate price-realized volatility time series in financial forecasting. A novel dynamic gamma process model adopted for realized volatility is integrated with traditional Bayesian dynamic linear models (DLMs) for asset price series. This represents reduced-form volatility leverage and feedback effects through use of realized volatility proxies in conditional DLMs for prices or returns, coupled with the synthesis of higher frequency data to track and anticipate volatility fluctuations. Analysis is computationally straightforward, extending conjugate-form Bayesian analyses for sequential filtering and model monitoring with simple and direct simulation for forecasting. A main applied setting is equity return forecasting with daily prices and realized volatility from high-frequency, intraday data. Detailed empirical studies of multiple S&P sector ETFs highlight the improvements achievable in asset price forecasting relative to standard models and deliver contextual insights on the nature and practical relevance of volatility leverage and feedback effects. The analytic structure and negligible extra computational cost will enable scaling to higher dimensions for multivariate price series forecasting for decouple/recouple portfolio construction and risk management applications.

Complexity vs Empirical Score

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

Why this score: The paper presents a sophisticated Bayesian framework utilizing dynamic gamma processes and conjugate state-space filtering, while also conducting detailed empirical backtesting on a collection of S&P sector ETFs. It successfully bridges theoretical analytic tractability with practical, applied high-frequency data modeling.

Research Flowchart

  flowchart TD
    A[Research Goal: Improve Asset Price Forecasting] --> B[Inputs: Daily Prices & Intraday Realized Volatility]
    B --> C[Method: Bayesian Dynamic Model Integration]
    C --> D[Processes: Sequential Filtering & Direct Simulation]
    D --> E[Analysis: Leverage & Feedback Effect Modeling]
    E --> F[Outcomes: Superior Forecasts & Scalable Risk Management]

Cited References (156)

  1. High-frequency realized stochastic volatility model (2024) — link
  2. Leverage and Volatility Feedback Effects in High-Frequency Data (2006) — link
  3. Forecasting Daily Volatility of Stock Price Index Using Daily Returns and Realized Volatility (2024) — link
  4. Bayesian emulation for multi-step optimization in decision problems (2019) — link
  5. Mixture models, M onte C arlo, B ayesian updating and dynamic models (1993) — link
  6. Modelling Financial Time Series (1986)
  7. Bayesian Analysis of Stochastic Volatility Models (1994) — link
  8. Statistical Aspects of ARCH and Stochastic Volatility (1996)
and 148 more
  1. Designing Realized Kernels to Measure the Ex-Post Variation of Equity Prices in the Presence of Noise (2008) — link
  2. Fractionally Integrated Generalized Autoregressive Conditional Heteroskedasticity (1996) — link
  3. The Analysis of Stochastic Volatility in the Presence of Daily Realized Measures (2013) — link
  4. New Frontiers for ARCH Models (2002) — link
  5. A Multiple Indicators Model for Volatility Using Intra-Daily Data (2006) — link
  6. Continuous Univariate Distributions, Volume 2 (1995)
  7. Realising the Future: Forecasting with High-Frequency-Based Volatility ( HEAVY ) Models (2010) — link
  8. Filtering via Simulation: Auxiliary Particle Filters (1999) — link
  9. Applied B ayesian Forecasting and Time Series Analysis (1994)
  10. Time Series: Modeling, Computation, and Inference (2021) — link
  11. The DYNAMICLINEAR Procedure (2024) — link
  12. The Matrix- F Prior for Estimating and Testing Covariance Matrices (2018) — link
  13. Fractional Differencing (1981) — link
  14. AN INTRODUCTION TO LONG-MEMORY TIME SERIES MODELS AND FRACTIONAL DIFFERENCING (1980) — link
  15. Long memory stochastic volatility : A bayesian approach (2000) — link
  16. Forecasting Daily Volatility of Stock Price Index Using Daily Returns and Realized Volatility (2024) — link
  17. Volatility and quantile forecasts by realized stochastic volatility models with generalized hyperbolic distribution (2016) — link
  18. Studies of Stock Price Volatility Changes (1976)
  19. Localized Realized Volatility Modelling (2010)
  20. Stochastic volatility model with leverage and asymmetrically heavy-tailed error using GH skew S tudent’s t-distribution (2012) — link
  21. Leverage, heavy-tails and correlated jumps in stochastic volatility models (2009) — link
  22. Stochastic volatility with leverage: Fast and efficient likelihood inference (2007) — link
  23. Measuring and Modeling Risk Using High-Frequency Data (2008) — link
  24. Realized Kernels in Practice: Trades and Quotes (2009) — link
  25. Combined Parameter and State Estimation in Simulation-Based Filtering (2001) — link
  26. Novel approach to nonlinear/non-Gaussian Bayesian state estimation (1993) — link
  27. The Stochastic Behavior of Common Stock Variances (1982)
  28. Measuring and Testing the Impact of News on Volatility (1993)
  29. Normal Discount B ayesian Models (1985)
  30. The Information Content of High-Frequency Data for Estimating Equity Return Models and Forecasting Risk (2010)
  31. The Economic Value of Realized Volatility: Using High-Frequency Returns for Option Valuation (2014)
  32. Leverage Effects in Stochastic Volatility with Time‐Varying Persistence (2011)
  33. Local Adaptive Multiplicative Error Models for High-Frequency Forecasts (2015) — link
  34. GENERALIZED AUTOREGRESSIVE SCORE MODELS WITH APPLICATIONS (2013) — link
  35. A multiple indicators model for volatility using intra-daily data (2006) — link
  36. On the Half-Cauchy Prior for a Global Scale Parameter (2012) — link
  37. Prior distributions for variance parameters in hierarchical models (Comment on article by Browne and Draper) (2006) — link
  38. Heterogeneous Information Arrivals and Return Volatility Dynamics: Uncovering the Long-Run in High Frequency Returns (1997) — link
  39. A Gamma Filter for Positive Parameter Estimation (2020) — link
  40. Bayesian Analysis of Stochastic Volatility Models (2002) — link
  41. Statistics for Long-Memory Processes (1994) — link
  42. Local scale models: State space alternative to integrated GARCH processes (1994) — link
  43. Realized GARCH: A Joint Model of Returns and Realized Measures of Volatility (2010) — link
  44. Realising the future: forecasting with high-frequency-based volatility (HEAVY) models (2010) — link
  45. Prior distributions for variance parameters in hierarchical models (comment on article by Browne and Draper) (2006) — link
  46. Bayesian forecasting of multivariate time series: S calability, structure uncertainty and decisions (with discussion) (2020) — link
  47. Post-processing posteriors over precision matrices to produce sparse graph estimates (2019) — link
  48. Decoupling shrinkage and selection in G aussian linear factor analysis (2024) — link
  49. Comparative politics and the synthetic control method (2015)
  50. Synthetic control methods for comparative case studies: E stimating the effect of C alifornia’s tobacco control program (2010)
  51. Compositional dynamic modelling for counterfactual prediction in multivariate time series (2024) — link
  52. Subjective intervention in formal models (1989)
  53. GPU -accelerated B ayesian learning and forecasting in simultaneous graphical dynamic linear models (2016) — link
  54. Bayesian forecasting and scalable multivariate volatility analysis using simultaneous graphical dynamic linear models (2017) — link
  55. Bayesian forecasting of stock Returns on the JSE using simultaneous graphical dynamic linear models — → 2307.08665
  56. Efficient computation of mean reverting portfolios using cyclical coordinate descent (2021) — link
  57. Model specification in multivariate time series (1989) — link
  58. Bayesian learning in sparse graphical factor models via annealed entropy (2010)
  59. Models of random sparse eigenmatrices and B ayesian analysis of multivariate structure (2016)
  60. Factor stochastic volatility with time varying loadings and M arkov switching regimes (2007)
  61. Parsimony inducing priors for large scale state–space models (2022) — link
  62. Time varying structural vector autoregressions and monetary policy (2005)
  63. Sparse B ayesian factor analysis when the number of factors is unknown (2024) — link
  64. Large time-varying parameter VARs (2013) — link
  65. Bayesian Multivariate Time Series Methods for Empirical Macroeconomics (2010) — link
  66. Large B ayesian vector autoregressions (2010)
  67. Bayesian dynamic factor models: L atent threshold approach (2013) — link
  68. Dynamics and sparsity in latent threshold factor models: A study in multivariate EEG signal processing (2017) — link
  69. Bayesian analysis of latent threshold dynamic models (2013) — link
  70. Dynamic network signal processing using latent threshold models (2015) — link
  71. High-dimensional sparse factor modelling– A pplications in gene expression genomics (2008)
  72. Comparative politics and the synthetic control method revisited: A note on Abadie et al. (2015) (2018) — link
  73. The inclusive synthetic control method (2020) — link
  74. An improved and extended B ayesian synthetic control (2021) — link
  75. An integrative analysis of cancer gene expression studies using B ayesian latent factor modelling (2009)
  76. Bayesian factor regression models in the \lq\lq large p, small n" paradigm (2003) — link
  77. Bayesian dynamic factor models and portfolio allocation (2000) — link
  78. Bayesian forecasting and portfolio decisions using dynamic dependent sparse factor models (2014) — link
  79. Spectral properties of matrices (2014) — link
  80. Variable prioritization in nonlinear black box methods: A genetic association case study (2019) — link
  81. Matrix Analysis (2012)
  82. System connectivity and matrix properties (1969) — link
  83. Spectra of Graphs : Theory and Application (1980) — link
  84. On low-rank directed acyclic graphs and causal structure learning (2024) — link
  85. On robustness of principal component regression (2021) — link
  86. Bayesian (non-)unique sparse factor modelling (2023)
  87. A B ayesian alternative to synthetic control for comparative case studies (2022)
  88. Cointegration and control: A ssessing the impact of events using time series data (2021) — link
  89. Inferring causal impact using Bayesian structural time-series models (2015) — link
  90. The economic costs of conflict: A case study of the Basque Country (2003)
  91. Estimating the effectiveness of permanent price reductions for competing products using multivariate Bayesian structural time series models (2022) — link
  92. Heterogeneous causal effects of neighbourhood policing in New York City with staggered adoption of the policy (2023) — link
  93. Dynamic graphical models with variable selection for effective connectivity (2023) — link
  94. SGDLM in python (2023)
  95. GPU -accelerated software for online learning of the simultaneous graphical DLM (2018)
  96. Experiments in stochastic computation for high-dimensional graphical models (2005) — link
  97. Covariance decomposition in undirected G aussian graphical models (2005) — link
  98. Adaptive variable selection for sequential prediction in multivariate dynamic models (2021) — link
  99. Shotgun stochastic search in regression with many predictors (2007) — link
  100. Bayesian predictive decision synthesis (2023) — link
  101. Predictive decision synthesis for portfolios: B etting on better models (2024) — link
  102. Bayesian computation in dynamic latent factor models (2022) — link
  103. Dynamic dependence networks: F inancial time series forecasting and portfolio decisions (with discussion) (2016) — link
  104. Generalized Autoregressive Conditional Heteroskedasticity (1986)
  105. A Capital Asset Pricing Model with Time-Varying Covariances (1988)
  106. Variance Risk Premiums (2009)
  107. Expected Stock Returns and Variance Risk Premia (2009)
  108. Intertemporal Asset Pricing without Consumption Data (1993)
  109. Power and Bipower Variation with Stochastic Volatility and Jumps (2004) — link
  110. Realized Variance and Market Microstructure Noise (2006) — link
  111. Time Series: Modeling, Computation, and Inference (2010)
  112. On Singular Wishart and Singular Multivariate Beta Distributions (1994) — link
  113. Aspects of Multivariate Statistical Theory (1982) — link
  114. Econometric Analysis of Realized Covariation: High Frequency Based Covariance, Regression, and Correlation in Financial Economics (2004) — link
  115. Dynamic Matrix-Variate Graphical Models (2007) — link
  116. Autoregressive Models for Variance Matrices: Stationary Inverse Wishart Processes (2011) — link
  117. Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk (1964) — link
  118. Security Prices, Risk, and Maximal Gains from Diversification (1965) — link
  119. Equilibrium in a Capital Asset Market (1966) — link
  120. Capital Market Equilibrium with Restricted Borrowing (1972) — link
  121. Portfolio Selection (1952) — link
  122. Infinite Divisibility of the Hyperbolic and Generalized Inverse Gaussian Distributions (1977) — link
  123. Statistical Properties of the Generalized Inverse Gaussian Distribution (1982) — link
  124. Distribution of the Ratio of Two Wishart Matrices and Cumulative Probability Evaluation by the Holonomic Gradient Method (2018) — link
  125. Dynamic Generalized Linear Models and B ayesian Forecasting (1985) — link
  126. A Tale of Two Time Scales: Determining Integrated Volatility with Noisy High-Frequency Data (2005) — link
  127. The Extreme Value Method for Estimating the Variance of the Rate of Return (1980) — link
  128. On the Estimation of Security Price Volatilities from Historical Data (1980) — link
  129. Estimating Variance from High, Low, and Closing Prices (1991)
  130. Drift-Independent Volatility Estimation Based on High, Low, Open and Close Prices (2000) — link
  131. Stochastic volatility: L ikelihood inference and comparison with ARCH models (1998) — link
  132. Monitoring and adaptation in B ayesian forecasting models (1986)
  133. Bayesian Forecasting and Dynamic Models (1997)
  134. Bayesian Analysis of Long Memory Time Series (1997)
  135. Exploiting the Errors: A Simple Approach for Improved Volatility Forecasting (2016) — link
  136. An Intertemporal Capital Asset Pricing Model (1973)
  137. Expected Stock Returns and Volatility (1987)
  138. Conditional Heteroskedasticity in Asset Returns: A New Approach (1991)
  139. On the Relation between the Expected Value and the Volatility of the Nominal Excess Return on Stocks (1993)
  140. Bayesian estimation of sparse dynamic factor models with order-independent and ex-post mode identification (2019) — link
  141. Multivariate B ayesian dynamic modeling for causal prediction: M ore on models, data and analyses (2024) — link
  142. Estimating Time Varying Risk Premia in the Term Structure: The ARCH-M Model (1987) — link
  143. The Scaled Beta2 Distribution as a Robust Prior for Scales (2017) — link
  144. Realized GARCH : A joint model for returns and realized measures of volatility (2012) — link
  145. Estimating stochastic volatility models using daily returns and realized volatility simultaneously (2009) — link
  146. A Simple Approximate Long-Memory Model of Realized Volatility (2009) — link
  147. Modeling and Forecasting Realized Volatility (2003) — link
  148. Econometric Analysis of Realized Volatility and Its Use in Estimating Stochastic Volatility Models (2002) — link

1 of 156 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.