Paper: arXiv 2510.19271

Authors: Jozef Barunik, Lukas Janasek, Attila Sarkany

Abstract

We study the dynamic investment decisions of investors who prioritise specific quantiles of outcomes over their expected values. Downside-focused agents targeting low quantiles reduce risk in states with high variance, while those with a preference for high quantiles concentrate in sleeves with high dispersion when there is potential for upside. These results provide a microfoundation for volatility management, demonstrating that reducing exposure in volatile states is an optimal response for risk-averse investors and rationalising inverse-variance heuristics. We propose a distributional actor-critic algorithm that learns time-consistent policies tailored to these specific risks, irrespective of the utilitys functional form. The quantile value can be mapped onto interpretable tilts, and the performance of empirically chosen portfolios aligns with investors objectives.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper provides a strong theoretical and mathematical foundation by integrating dynamic quantile preferences with distributional reinforcement learning to microfound volatility-managed strategies. It maintains high empirical rigor through multi-asset dynamic portfolio backtests and offers exceptional reproducibility by providing an official public GitHub repository for its code.

Research Flowchart

  flowchart TD
    A["Research Question: How do quantile-targeted preferences impact dynamic portfolio choice?"] --> B["Inputs: Cross-Asset Market Data & Investor Risk Profiles"]
    B --> C["Methodology: Distributional Actor-Critic Algorithm"]
    C --> D["Process: Learn Time-Consistent Policies across Expected & Extreme Quantiles"]
    D --> E["Finding 1: Risk-averse down-quantile targeting provides microfoundation for volatility management"]
    D --> F["Finding 2: High-quantile preference concentrates capital in high-dispersion upside sleeves"]

Cited References (102)

  1. On the Performance of Volatility-Managed Portfolios (2020)
  2. Substitution, Risk Aversion, and the Temporal Behavior of Consumption and Asset Returns: A Theoretical Framework (1989)
  3. Substitution, Risk Aversion, and the Temporal Behavior of Consumption and Asset Returns: An Empirical Analysis (1991)
  4. Measuring Fund Strategy and Performance in Changing Economic Conditions (1996) — link
  5. Conditioning Variables and the Cross Section of Stock Returns (1999) — link
  6. Conditional skewness in asset pricing tests (2000)
  7. Tail risk and asset prices (2014)
  8. Portfolio Choice and Equilibrium in Capital Markets with Safety-First Investors (1977) — link
and 94 more
  1. Safety First and the Holding of Assets (1952) — link
  2. A Characterization of the Distributions That Imply Mean-Variance Utility Functions (1983) — link
  3. Choosing factors (2018)
  4. A General Equilibrium Model of Portfolio Insurance (1995) — link
  5. Value-at-Risk-Based Risk Management: Optimal Policies and Asset Prices (2001) — link
  6. Liability-Driven Investment with Downside Risk (2013) — link
  7. Downside Risk (2006) — link
  8. Conditional Skewness in Asset Pricing Tests (2000)
  9. Unmasking Mutual Fund Derivative Use (2025) — link
  10. The Role of Taxes in the Rise of ETFs (2025) — link
  11. The Variation of Certain Speculative Prices (1963)
  12. The Behavior of Stock-Market Prices (1965)
  13. Empirical Properties of Asset Returns: Stylized Facts and Statistical Issues (2001)
  14. Autoregressive Conditional Skewness (1999)
  15. Mental Accounting, Loss Aversion, and Individual Stock Returns (2001)
  16. Asset Allocation under Multivariate Regime Switching (2007)
  17. Asymmetries and Portfolio Choice (2017) — link
  18. Portfolios for Long-Term Investors (2022) — link
  19. Volatility-Managed Portfolios (2017) — link
  20. Household portfolio underdiversification and probability weighting: Evidence from the field (2021)
  21. Cumulative prospect theory, option returns, and the variance premium (2019)
  22. What drives the disposition effect? An analysis of a long-standing preference-based explanation (2009)
  23. Prospect theory and stock returns: An empirical test (2016)
  24. Quantile Approach to Intertemporal Consumption with Multiple Assets (2023)
  25. Pessimistic portfolio allocation and Choquet expected utility (2004)
  26. A Dynamic Model of Private Asset Allocation — → 2503.01099
  27. A comprehensive machine learning framework for dynamic portfolio choice with transaction costs (2023)
  28. Generalized Disappointment Aversion and Asset Prices (2010)
  29. Stochastic approximation: a dynamical systems viewpoint (2008)
  30. International asset allocation with regime shifts (2002)
  31. The smart beta mirage (2024)
  32. When passive funds affect prices: evidence from volatility and commodity ETFs (2024)
  33. Competition for attention in the ETF space (2023)
  34. Tradable risk factors for institutional and retail investors (2025)
  35. Stock market prices do not follow random walks: Evidence from a simple specification test (1988)
  36. Generalized autoregressive conditional heteroskedasticity (1986)
  37. A Multifactor Perspective on Volatility-Managed Portfolios (2024)
  38. Approximation capabilities of multilayer feedforward networks (1991)
  39. Dynamic economics with quantile preferences (2025)
  40. An axiomatization of quantiles on the domain of distribution functions (2009)
  41. Gradient-Based Reinforcement Learning for Dynamic Quantile (2025)
  42. Gradient-based reinforcement learning for dynamic quantile models (2026) — link
  43. Portfolio Selection (1952) — link
  44. How regimes affect asset allocation (2004)
  45. Asset allocation under multivariate regime switching (2007)
  46. Dynamic portfolio optimization with inverse covariance clustering (2023)
  47. An online portfolio selection algorithm using clustering approaches and considering transaction costs (2020)
  48. Dynamic portfolio optimization using technical analysis-based clustering (2022)
  49. Dynamic asset allocation strategy using a state-dependent Markov model: Applications to international equity markets (2022)
  50. Dynamic portfolio optimization across hidden market regimes (2018)
  51. Reinforcement learning: An introduction (2018)
  52. Deep reinforcement learning for portfolio management (2023)
  53. Continuous control with deep reinforcement learning (2015) — link
  54. Reinforcement-learning based portfolio management with augmented asset movement prediction states (2020)
  55. A deep Q-learning portfolio management framework for the cryptocurrency market (2020)
  56. Worst cases policy gradients (2019) — link
  57. Risk-sensitive policies for portfolio management (2022)
  58. A five-factor asset pricing model (2015)
  59. \ldots and the cross-section of expected returns (2016)
  60. Taming the factor zoo: A test of new factors (2020)
  61. Factor Investing and Risk Management: Is Smart-Beta Diversification Smart? (2021)
  62. Optimal blending of smart beta and multifactor portfolios (2018)
  63. A data-driven neural network approach to dynamic factor investing
  64. Dirichlet policies for reinforced factor portfolios (2020) — link
  65. Dsac: Distributional soft actor critic for risk-sensitive reinforcement learning (2020) — link
  66. Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor (2018)
  67. Distributional reinforcement learning with quantile regression (2018)
  68. A novel dynamic asset allocation system using Feature Saliency Hidden Markov models for smart beta investing (2021)
  69. Layer normalization (2016) — link
  70. Practical deep reinforcement learning approach for stock trading (2018)
  71. Dynamic quantile models of rational behavior (2019)
  72. Ordinal utility models of decision making under uncertainty (1988)
  73. Quantile maximization in decision theory (2010)
  74. Quantile-based policy optimization for reinforcement learning (2022)
  75. Lifetime portfolio selection under uncertainty: The continuous-time case (1969)
  76. Reinforcement learning for sequential decision and optimal control (2023)
  77. Learning from delayed rewards (1989)
  78. A distributional perspective on reinforcement learning (2017)
  79. Machine learning for continuous-time finance (2024)
  80. Dynamic variational preferences (2006)
  81. Actor–Critic or Critic–Actor? A Tale of Two Time Scales (2023)
  82. Finite sample analysis of two-timescale stochastic approximation with applications to reinforcement learning (2018)
  83. Implicit quantile networks for distributional reinforcement learning (2018)
  84. Discovering and removing exogenous state variables and rewards for reinforcement learning (2018)
  85. Neurocomputing: Foundations of research (1988)
  86. What matters in on-policy reinforcement learning? a large-scale empirical study (2020) — link
  87. The liquidity of active ETFs (2021)
  88. Advances in prospect theory: Cumulative representation of uncertainty (1992)
  89. A class of distortion operators for pricing financial and insurance risks (2000)
  90. Portfolio selection in quantile decision models (2022)
  91. A unified approach to interpreting model predictions (2017)
  92. A value for n-person games (1953)
  93. Optimum consumption and portfolio rules in a continuous-time model (1971)
  94. Reinforcement learning with exogenous states and rewards (2023) — link

1 of 102 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.