Paper: arXiv 2412.03668

Abstract

In this paper we develop a novel hidden Markov graphical model to investigate time-varying interconnectedness between different financial markets. To identify conditional correlation structures under varying market conditions and accommodate stylized facts embedded in financial time series, we rely upon the generalized hyperbolic family of distributions with time-dependent parameters evolving according to a latent Markov chain. We exploit its location-scale mixture representation to build a penalized EM algorithm for estimating the state-specific sparse precision matrices by means of an $L_1$ penalty. The proposed approach leads to regime-specific conditional correlation graphs that allow us to identify different degrees of network connectivity of returns over time. The methodology’s effectiveness is validated through simulation exercises under different scenarios. In the empirical analysis we apply our model to daily returns of a large set of market indexes, cryptocurrencies and commodity futures over the period 2017-2023.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 7.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced statistical techniques including generalized hyperbolic distributions, location-scale mixture representations, and penalized EM algorithms, demonstrating high mathematical complexity. It validates the methodology through simulation studies and applies it to real-world financial data (indices, cryptocurrencies, commodities) over a defined period, showing strong empirical rigor.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Model Time-varying Market Interconnectedness<br>under Dynamic, Heavy-tailed Conditions"] --> B["Methodology: Hidden Markov Graphical Model<br>with State-dependent Generalized Hyperbolic Distributions"]
  B --> C["Computational Process<br>Penalized EM Algorithm<br>(L1 Penalty for Sparse Precision Matrices)"]
  C --> D["Simulation Validation<br>Test Model Recovery under Various Scenarios"]
  C --> E["Empirical Application<br>Daily Returns (2017-2023)<br>Indices, Crypto, Commodities"]
  D --> F["Key Outcomes"]
  E --> F
  F --> G["Regime-specific Conditional Correlation Graphs"]
  F --> H["Identification of Time-varying Network Connectivity"]

Cited References (148)

  1. CAV ia R : conditional autoregressive value at risk by regression quantiles (2004)
  2. A comparison of generalized hyperbolic distribution models for equity returns (2014)
  3. Estimating the tails of loss severity via conditional risk measures for the family of symmetric generalised hyperbolic distributions (2015)
  4. Modeling heavy-tailed stock index returns using the generalized hyperbolic distribution (2009)
  5. Nonparametric risk management with generalized hyperbolic distributions (2008)
  6. Time-varying beta risk of P an- E uropean industry portfolios: A comparison of alternative modeling techniques (2008)
  7. Expectile hidden M arkov regression models for analyzing cryptocurrency returns — → 2301.09722
  8. Quantile and expectile copula-based hidden M arkov regression models for the analysis of the cryptocurrency market — → 2307.06400
and 140 more
  1. The generalised hyperbolic distribution and its subclass in the analysis of a new era of cryptocurrencies: Ethereum and its financial risk (2019)
  2. A mixture of generalized hyperbolic distributions (2015)
  3. A dynamic analysis of stock markets using a hidden M arkov model (2013)
  4. Long memory of financial time series and hidden M arkov models with time-varying parameters (2017)
  5. Hidden M arkov and semi- M arkov models with multivariate leptokurtic-normal components for robust modeling of daily returns series (2019)
  6. Exploring the dependencies among main cryptocurrency log-returns: A hidden M arkov model (2022)
  7. Hidden M arkov and other models for discrete-valued time series (1997)
  8. Comparing partitions (1985)
  9. Penalized estimation in high-dimensional hidden M arkov models with state-specific graphical models (2013)
  10. On fitting mixture models (1999)
  11. Estimating the dimension of a model (1978)
  12. Distribution of the estimators for autoregressive time series with a unit root (1979)
  13. Confidence intervals for hidden M arkov model parameters (2000)
  14. The EM algorithm and extensions (2007)
  15. The ECME algorithm: a simple extension of EM and ECM with faster monotone convergence (1994)
  16. R: A Language and Environment for Statistical Computing (2023) — link
  17. Hidden M arkov models with nonelliptically contoured state densities (2010)
  18. Herding behaviour in cryptocurrencies (2019)
  19. A maximization technique occurring in the statistical analysis of probabilistic functions of M arkov chains (1970)
  20. The generalized hyperbolic family and automatic model selection through the multiple-choice LASSO (2023) — link
  21. Quantitative risk management: concepts, techniques and tools-revised edition (2015)
  22. Using time-series and sentiment analysis to detect the determinants of B itcoin prices (2015)
  23. What are the main drivers of the Bitcoin p rice? E vidence from wavelet coherence analysis (2015)
  24. A model of financialization of commodities (2016)
  25. Bitcoin, gold and the dollar–A GARCH volatility analysis (2016)
  26. Quantile Connectedness: modelling tail behaviour in the topology of financial networks (2018)
  27. Bitcoin: Medium of exchange or speculative assets? (2018)
  28. Co-explosivity in the cryptocurrency market (2019)
  29. Cryptocurrencies as hedges and safe-havens for US equity sectors (2020)
  30. Bitcoin, gold, and commodities as safe havens for stocks: New insight through wavelet analysis (2020)
  31. A rational expectations model of financial contagion (2002)
  32. Coronavirus (COVID-19)-An epidemic or pandemic for financial markets (2020)
  33. Financial markets under the global pandemic of COVID-19 (2020)
  34. On the hedge and safe haven properties of B itcoin: Is it really more than a diversifier? (2017)
  35. Network causality structures among B itcoin and other financial assets: A directed acyclic graph approach (2018)
  36. Crypto price discovery through correlation networks (2021)
  37. Lead behaviour in b itcoin markets (2020)
  38. Is Bitcoin a real currency? An economic appraisal (2015)
  39. Virtual currency schemes-a further analysis (2015)
  40. An analysis of the literature on systemic financial risk: A survey (2017)
  41. Global Financial Stability Report: Responding to the Financial Crisis and Measuring Systemic Risks (2009)
  42. Economic networks: The new challenges (2009)
  43. CoRisk: credit risk contagion with correlation network models (2018)
  44. Bank lending and systemic risk: A financial-real sector network approach with feedback (2018)
  45. Interconnectedness in the interbank market (2019)
  46. Axioms for centrality (2014)
  47. Dynamic asset trees and Black Monday (2003)
  48. Dynamics of market correlations: Taxonomy and portfolio analysis (2003)
  49. High-dimensional semiparametric G aussian copula graphical models (2012)
  50. Regularized rank-based estimation of high-dimensional nonparanormal graphical models (2012)
  51. The nonparanormal: Semiparametric estimation of high dimensional undirected graphs. (2009)
  52. Tiger: A tuning-insensitive approach for optimally estimating gaussian graphical models (2017)
  53. A SINful approach to G aussian graphical model selection (2008)
  54. Model selection for G aussian concentration graphs (2004)
  55. Covariance selection (1972)
  56. A constrained $\ell_1$ minimization approach to sparse precision matrix estimation (2011)
  57. Quantile graphical models: prediction and conditional independence with applications to systemic risk (2016) — link
  58. The cross-quantilogram: Measuring quantile dependence and testing directional predictability between time series (2016)
  59. Financial bubbles: a study of co-explosivity in the cryptocurrency market (2020)
  60. On use of the EM algorithm for penalized likelihood estimation (1990)
  61. Estimating global bank network connectedness (2018)
  62. Financial systemic risk measurement based on causal network connectedness analysis (2019)
  63. Exploring the dynamic relationships between cryptocurrencies and other financial assets (2018)
  64. Evolution of the bitcoin address graph (2017)
  65. Robust graphical modeling of gene networks using classical and alternative t-distributions (2011)
  66. Extended B ayesian information criteria for model selection with large model spaces (2008)
  67. High-dimensional I sing model selection with B ayesian information criteria (2015)
  68. Extended Bayesian information criteria for G aussian graphical models (2010)
  69. Mixtures of shifted asymmetriclaplace distributions (2013)
  70. Spillovers and Directional Predictability with a Cross - Quantilogram Analysis : The Case of U . S . and Chinese Agricultural Futures (2016) — link
  71. Asymmetric Risk Spillovers between Oil and Agricultural Commodities (2018) — link
  72. Connectedness of financial institutions in E urope: A network approach across quantiles (2020)
  73. Graphical models (1996)
  74. Regression Q uantiles (1978)
  75. Quantile R egression (2005)
  76. Conditional risk based on multivariate hazard scenarios (2018)
  77. On multivariate extensions of the conditional V alue-at- R isk measure (2015)
  78. A directional multivariate V alue at R isk (2015)
  79. Value at R isk, 3rd Ed.: The New Benchmark for Managing Financial Risk (2006)
  80. The sparse method of simulated quantiles: An application to portfolio optimization (2018)
  81. Market Risk Analysis, V alue at R isk M odels (2009)
  82. A M arkov regime switching approach for hedging energy commodities (2008)
  83. Quantile regression for longitudinal data (2004)
  84. Bayesian quantile regression (2001)
  85. Spatial quantile multiple regression using the asymmetric L aplace process (2012)
  86. On multivariate quantile regression (2003)
  87. VAR for V a R : Measuring tail dependence using multivariate regression quantiles (2015)
  88. MULTIVARIATE QUANTILES AND MULTIPLE-OUTPUT REGRESSION QUANTILES: FROM L $_ 1 $ OPTIMIZATION TO HALFSPACE DEPTH (2010)
  89. M-quantile regression for multivariate longitudinal data: analysis of the M illennium C ohort S tudy data (2016) — link
  90. Markov regime-switching quantile regression models and financial contagion detection (2016)
  91. The L aplace distribution and generalizations: a revisit with applications to communications, economics, engineering, and finance (2012)
  92. Multiple Risk Measures for Multivariate Dynamic Heavy-Tailed Models (2017)
  93. Conditional Risk Based on Multivariate Hazard Scenarios (2016)
  94. Multivariate M arkov- S witching models and tail risk interdependence (2013) — link
  95. C o V a R (2016)
  96. Joint estimation of conditional quantiles in multivariate linear regression models with an application to financial distress (2019)
  97. Mixed H idden M arkov models: an extension of the H idden M arkov model to the longitudinal data setting (2007)
  98. Latent M arkov models for longitudinal data (2012)
  99. Hidden M arkov models for time series: an introduction using R (2016)
  100. Mixed H idden M arkov quantile regression models for longitudinal data with possibly incomplete sequences (2018)
  101. Mixed hidden M arkov models for quantiles (2014)
  102. Generalized autoregressive score models with applications (2013)
  103. A non-homogeneous hidden M arkov model for precipitation occurrence (1999)
  104. Quantile regression for longitudinal data based on latent M arkov subject-specific parameters (2012)
  105. Maximum likelihood from incomplete data via the EM algorithm (1977)
  106. The adaptive lasso and its oracle properties (2006)
  107. Variable selection via nonconcave penalized likelihood and its oracle properties (2001)
  108. Nearly unbiased variable selection under minimax concave penalty (2010)
  109. Variable selection in robust regression models for longitudinal data (2012)
  110. Regularized partially functional quantile regression (2017)
  111. Quantile regression for longitudinal data using the asymmetric L aplace distribution (2006)
  112. Mixed-effects models for conditional quantiles with longitudinal data (2009)
  113. Numerical maximisation of likelihood: A neglected alternative to EM ? (2014)
  114. Regression Shrinkage and Selection via the L asso (1996)
  115. Mixed H idden M arkov models for longitudinal data: an overview (2011)
  116. Computational issues in parameter estimation for stationary hidden M arkov models (2008)
  117. Quantiles for counts (2005)
  118. Likelihood inference for a class of latent M arkov models under linear hypotheses on the transition probabilities (2006)
  119. Linear mixed models: a practical guide using statistical software (2014)
  120. Nonparametric maximum likelihood estimation of a mixing distribution (1978)
  121. Cross- C ountry A ssessment of S ystemic R isk in the E uropean S tock M arket: E vidence from a C o V a R A nalysis (2018) — link
  122. Selection of V alue at R isk models for energy commodities (2018)
  123. Regularization and variable selection via the elastic net (2005)
  124. The geometry of mixture likelihoods: a general theory (1983)
  125. A multivariate extension of the dynamic logit model for longitudinal data based on a latent M arkov heterogeneity structure (2009)
  126. Linear quantile mixed models (2014)
  127. Hidden M arkov M odels and the B aum- W elch algorithm (2003)
  128. Handbook of Q uantile R egression (2017)
  129. Longitudinal analysis of the strengths and difficulties questionnaire scores of the M illennium C ohort S tudy children in E ngland using M -quantile random-effects regression (2016)
  130. Model selection through sparse maximum likelihood estimation for multivariate G aussian or binary data (2008)
  131. Sparse inverse covariance estimation with the graphical L asso (2008)
  132. Forecasting V a R and ES using a joint quantile regression and its implications in portfolio allocation (2021)
  133. Quantile M ixed H idden M arkov M odels for multivariate longitudinal data: an application to children’s strengths and difficulties questionnaire scores (2022) — link
  134. Bayesian Tail Risk Interdependence Using Quantile Regression (2015)
  135. The structure and degree of dependence: A quantile regression approach (2013)
  136. Model selection and estimation in the G aussian graphical model (2007)
  137. High-dimensional graphs and variable selection with the lasso (2006)
  138. Estimation of symmetry-constrained G aussian graphical models: application to clustered dense networks (2015)
  139. Modeling systemic risk with Markov switching graphical SUR models (2019)
  140. The L aplace distribution and generalizations: a revisit with applications to communications, economics, engineering, and finance (2001)

2 of 148 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.