Yang Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yang Liu has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Zongxia Liang
Department of Mathematical Sciences, Tsinghua University, Beijing 100084, China
7 shared publications - Ruodu Wang
Department of Statistics and Actuarial Science, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada
6 shared publications - Lin He
School of Finance, Renmin University of China, Beijing, China
3 shared publications - Peng Liu
School of Mathematics, Statistics and Actuarial Science, University of Essex, Colchester, UK
3 shared publications - Tolulope Fadina
Department of Mathematics, University of Illinois Urbana-Champaign, Champaign, USA
3 shared publications Bingzhen Geng
School of Big data and Statistics, Anhui University, Hefei, Anhui, 230601, China
2 shared publications
Latest Publications
- 5 October 2026
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Financial Mathematics
- Vol. 17 (4) , 1533-1580
Value-at-Risk- and Expectile-Based Systemic Risk Measures and Second-Order Asymptotics: With Applications to Diversification
Systemic risk measures play a crucial role in analyzing individual losses conditional on extreme system-wide disasters. In this paper, we provide a unified asymptotic treatment for systemic risk measures. First, we classify them into two families of Value-at-Risk- (VaR-) and expectile-based systemic risk measures. While VaR-based risk measures have been extensively studied, in the latter family, we propose two new systemic risk measures, named the Individual Conditional Expectile (ICE) and the Systemic Individual Conditional Expectile (SICE), as alternatives to Marginal Expected Shortfall (MES) and Systemic Expected Shortfall (SES). Second, to characterize general mutually dependent and heavy-tailed risks, we consider a multivariate loss system following a multivariate Sarmanov distribution with common marginal distributions exhibiting second-order regular variation. Third, within this setting, we provide second-order asymptotic results for both families of systemic risk measures. These results extend standard first-order asymptotics and allow for more accurate tail approximations. Through numerical and analytical examples, we demonstrate the superiority of second-order asymptotics in accurately assessing systemic risk. We further conduct a comprehensive comparison between expectile-based and VaR-based systemic risk measures. The results indicate that expectile-based measures often yield higher asymptotic accuracy than VaR-based ones, emphasizing the former’s potential advantages in reporting extreme events and tail risk. As a financial application, we use the asymptotic treatment to discuss the diversification benefits associated with various risk measures. Finally, we extend and obtain the second-order asymptotic formulas for generalized-quantile-based systemic risk measures with power functions.
- 5 October 2026
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Financial Mathematics
- Vol. 17 (4) , 1533-1580
Value-at-Risk- and Expectile-Based Systemic Risk Measures and Second-Order Asymptotics: With Applications to Diversification
Systemic risk measures play a crucial role in analyzing individual losses conditional on extreme system-wide disasters. In this paper, we provide a unified asymptotic treatment for systemic risk measures. First, we classify them into two families of Value-at-Risk- (VaR-) and expectile-based systemic risk measures. While VaR-based risk measures have been extensively studied, in the latter family, we propose two new systemic risk measures, named the Individual Conditional Expectile (ICE) and the Systemic Individual Conditional Expectile (SICE), as alternatives to Marginal Expected Shortfall (MES) and Systemic Expected Shortfall (SES). Second, to characterize general mutually dependent and heavy-tailed risks, we consider a multivariate loss system following a multivariate Sarmanov distribution with common marginal distributions exhibiting second-order regular variation. Third, within this setting, we provide second-order asymptotic results for both families of systemic risk measures. These results extend standard first-order asymptotics and allow for more accurate tail approximations. Through numerical and analytical examples, we demonstrate the superiority of second-order asymptotics in accurately assessing systemic risk. We further conduct a comprehensive comparison between expectile-based and VaR-based systemic risk measures. The results indicate that expectile-based measures often yield higher asymptotic accuracy than VaR-based ones, emphasizing the former’s potential advantages in reporting extreme events and tail risk. As a financial application, we use the asymptotic treatment to discuss the diversification benefits associated with various risk measures. Finally, we extend and obtain the second-order asymptotic formulas for generalized-quantile-based systemic risk measures with power functions.
- Open Access
- 25 August 2026
- Published by Elsevier in SSRN Electronic Journal
Behavioral Participating Insurance: Optimal Investment under Probability Distortion and Aspiration Constraints
We study optimal investment for insurers managing participating (profit-sharing) contracts under probability distortion and probability benchmark (aspiration) constraints. The problem combines three theoretical complexities: (i) nonconcave effective utilities induced by embedded guarantees and surplus-sharing rules, (ii) probability weighting capturing behavioral aspects of long-horizon decisions, and (iii) aspiration-type constraints formalizing solvency requirements. Using quantile formulations and concavification techniques, we derive explicit closed-form solutions for optimal terminal wealth and trading strategies in both complete and incomplete Black-Scholes markets. Our utility class accommodates the piecewise hyperbolic absolute risk aversion (PHARA) family and covers nonconcavities arising naturally in insurance contexts. The framework reveals how probability distortion weakens lock-in behavior and induces time inconsistency: under inverse S-shaped distortions, insurers overestimate upside probabilities and increase risky investment relative to undistorted benchmarks. Asymptotic analysis and numerical illustrations demonstrate regime switches in optimal policies driven by regulatory thresholds and capital constraints. Our results extend the hope-fear-aspirations framework of He and Zhou (2016) and provide practical insights for managing insurance balance sheets under behavioral preferences and solvency constraints.
- 25 August 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessBehavioral Participating Insurance: Optimal Investment under Probability Distortion and Aspiration Constraints
We study optimal investment for insurers managing participating (profit-sharing) contracts under probability distortion and probability benchmark (aspiration) constraints. The problem combines three theoretical complexities: (i) nonconcave effective utilities induced by embedded guarantees and surplus-sharing rules, (ii) probability weighting capturing behavioral aspects of long-horizon decisions, and (iii) aspiration-type constraints formalizing solvency requirements. Using quantile formulations and concavification techniques, we derive explicit closed-form solutions for optimal terminal wealth and trading strategies in both complete and incomplete Black-Scholes markets. Our utility class accommodates the piecewise hyperbolic absolute risk aversion (PHARA) family and covers nonconcavities arising naturally in insurance contexts. The framework reveals how probability distortion weakens lock-in behavior and induces time inconsistency: under inverse S-shaped distortions, insurers overestimate upside probabilities and increase risky investment relative to undistorted benchmarks. Asymptotic analysis and numerical illustrations demonstrate regime switches in optimal policies driven by regulatory thresholds and capital constraints. Our results extend the hope-fear-aspirations framework of He and Zhou (2016) and provide practical insights for managing insurance balance sheets under behavioral preferences and solvency constraints.
- 14 July 2026
- Published by Springer Nature in Science China Mathematics
- p. 1-22
Second-order asymptotic analysis of tail probabilities of bidimensional randomly weighted sums
- 14 July 2026
- Published by Springer Nature in Science China Mathematics
- p. 1-22
Second-order asymptotic analysis of tail probabilities of bidimensional randomly weighted sums
- Open Access
Robust Bayesian Dynamic Programming for On-policy Risk-sensitive Reinforcement Learning
We propose a novel framework for risk-sensitive reinforcement learning (RSRL) that incorporates robustness against transition uncertainty. We define two distinct yet coupled risk measures: an inner risk measure addressing state and cost randomness and an outer risk measure capturing transition dynamics uncertainty. Our framework unifies and generalizes most existing RL frameworks by permitting general coherent risk measures for both inner and outer risk measures. Within this framework, we construct a risk-sensitive robust Markov decision process (RSRMDP), derive its Bellman equation, and provide error analysis under a given posterior distribution. We further develop a Bayesian Dynamic Programming (Bayesian DP) algorithm that alternates between posterior updates and value iteration. The approach employs an estimator for the risk-based Bellman operator that combines Monte Carlo sampling with convex optimization, for which we prove strong consistency guarantees. Furthermore, we demonstrate that the algorithm converges to a near-optimal policy in the training environment and analyze both the sample complexity and the computational complexity under the Dirichlet posterior and CVaR. Finally, we validate our approach through two numerical experiments. The results exhibit excellent convergence properties while providing intuitive demonstrations of its advantages in both risk-sensitivity and robustness. Empirically, we further demonstrate the advantages of the proposed algorithm through an application on option hedging.
Open AccessRobust Bayesian Dynamic Programming for On-policy Risk-sensitive Reinforcement Learning
We propose a novel framework for risk-sensitive reinforcement learning (RSRL) that incorporates robustness against transition uncertainty. We define two distinct yet coupled risk measures: an inner risk measure addressing state and cost randomness and an outer risk measure capturing transition dynamics uncertainty. Our framework unifies and generalizes most existing RL frameworks by permitting general coherent risk measures for both inner and outer risk measures. Within this framework, we construct a risk-sensitive robust Markov decision process (RSRMDP), derive its Bellman equation, and provide error analysis under a given posterior distribution. We further develop a Bayesian Dynamic Programming (Bayesian DP) algorithm that alternates between posterior updates and value iteration. The approach employs an estimator for the risk-based Bellman operator that combines Monte Carlo sampling with convex optimization, for which we prove strong consistency guarantees. Furthermore, we demonstrate that the algorithm converges to a near-optimal policy in the training environment and analyze both the sample complexity and the computational complexity under the Dirichlet posterior and CVaR. Finally, we validate our approach through two numerical experiments. The results exhibit excellent convergence properties while providing intuitive demonstrations of its advantages in both risk-sensitivity and robustness. Empirically, we further demonstrate the advantages of the proposed algorithm through an application on option hedging.
- 11 December 2025
- Published by Elsevier in Insurance: Mathematics and Economics
- Vol. 126
Asymptotics of systemic risk in a renewal model with multiple business lines and heterogeneous claims
- 11 December 2025
- Published by Elsevier in Insurance: Mathematics and Economics
- Vol. 126
Asymptotics of systemic risk in a renewal model with multiple business lines and heterogeneous claims
- Open Access
Bayesian Distributionally Robust Merton Problem with Nonlinear Wasserstein Projections
We revisit Merton's continuous-time portfolio selection through a data-driven, distributionally robust lens. Our aim is to tap the benefits of frequent trading over short horizons while acknowledging that drift is hard to pin down, whereas volatility can be screened using realized or implied measures for appropriately selected assets. Rather than time-rectangular distributional robust control -- which replenishes adversarial power at every instant and induces over-pessimism -- we place a single ambiguity set on the drift prior within a Bayesian Merton model. This prior-level ambiguity preserves learning and tractability: a minimax swap reduces the robust control to optimizing a nonlinear functional of the prior, enabling Karatzas and Zhao \cite{KZ98}-type's closed-form evaluation for each candidate prior. We then characterize small-radius worst-case priors under Wasserstein uncertainty via an explicit asymptotically optimal pushforward of the nominal prior, and we calibrate the ambiguity radius through a nonlinear Wasserstein projection tailored to the Merton functional. Synthetic and real-data studies demonstrate reduced pessimism relative to DRC and improved performance over myopic DRO-Markowitz under frequent rebalancing.
Open AccessBayesian Distributionally Robust Merton Problem with Nonlinear Wasserstein Projections
We revisit Merton's continuous-time portfolio selection through a data-driven, distributionally robust lens. Our aim is to tap the benefits of frequent trading over short horizons while acknowledging that drift is hard to pin down, whereas volatility can be screened using realized or implied measures for appropriately selected assets. Rather than time-rectangular distributional robust control -- which replenishes adversarial power at every instant and induces over-pessimism -- we place a single ambiguity set on the drift prior within a Bayesian Merton model. This prior-level ambiguity preserves learning and tractability: a minimax swap reduces the robust control to optimizing a nonlinear functional of the prior, enabling Karatzas and Zhao \cite{KZ98}-type's closed-form evaluation for each candidate prior. We then characterize small-radius worst-case priors under Wasserstein uncertainty via an explicit asymptotically optimal pushforward of the nominal prior, and we calibrate the ambiguity radius through a nonlinear Wasserstein projection tailored to the Merton functional. Synthetic and real-data studies demonstrate reduced pessimism relative to DRC and improved performance over myopic DRO-Markowitz under frequent rebalancing.
- Open Access
Extended Convolution Bounds on the Fréchet Problem: Robust Risk Aggregation and Risk Sharing
In this paper, we provide extended convolution bounds for the Fréchet problem and discuss related implications in quantitative risk management. First, we establish a new form of inequality for the Range-Value-at-Risk (RVaR). Based on this inequality, we obtain bounds for robust risk aggregation with dependence uncertainty for (i) RVaR, (ii) inter-RVaR difference and (iii) inter-quantile difference, and provide sharpness conditions. These bounds are called extended convolution bounds, which not only complement the results in the literature (convolution bounds in Blanchet et al. (2025)) but also offer results for some variability measures. Next, applying the above inequality, we study the risk sharing for the averaged quantiles (corresponding to risk sharing for distortion risk measures with special inverse S-shaped distortion functions), which is a non-convex optimization problem. We obtain the expression of the minimal value of the risk sharing and the explicit expression for the corresponding optimal allocation, which is comonotonic risk sharing for large losses and counter-comonotonic risk sharing for small losses or large gains. Finally, we explore the dependence structure for the optimal allocations, showing that the optimal allocation does not exist if the risk is not bounded from above.
Open AccessExtended Convolution Bounds on the Fréchet Problem: Robust Risk Aggregation and Risk Sharing
In this paper, we provide extended convolution bounds for the Fréchet problem and discuss related implications in quantitative risk management. First, we establish a new form of inequality for the Range-Value-at-Risk (RVaR). Based on this inequality, we obtain bounds for robust risk aggregation with dependence uncertainty for (i) RVaR, (ii) inter-RVaR difference and (iii) inter-quantile difference, and provide sharpness conditions. These bounds are called extended convolution bounds, which not only complement the results in the literature (convolution bounds in Blanchet et al. (2025)) but also offer results for some variability measures. Next, applying the above inequality, we study the risk sharing for the averaged quantiles (corresponding to risk sharing for distortion risk measures with special inverse S-shaped distortion functions), which is a non-convex optimization problem. We obtain the expression of the minimal value of the risk sharing and the explicit expression for the corresponding optimal allocation, which is comonotonic risk sharing for large losses and counter-comonotonic risk sharing for small losses or large gains. Finally, we explore the dependence structure for the optimal allocations, showing that the optimal allocation does not exist if the risk is not bounded from above.
- 1 September 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 73 (5) , 2761-2781
Convolution Bounds on Quantile Aggregation
Advancing Risk Assessment: New Ways To Compute Quantile Aggregation This issue features a pivotal study on quantile aggregation amid dependence uncertainty, an area critical to finance, risk management, and statistics. The authors introduce “convolution bounds,” derived from a recent inf-convolution formula of quantiles and related risk measures. The obtained analytical tools unify existing results and enhance the understanding of quantile methods by providing general, sharp, and computationally efficient solutions. The results offer insights into the extremal dependence structures, with several implications in risk management and economic analysis applications. For more detailed insights, read the full paper, “Convolution Bounds on Quantile Aggregation” (reference: [insert reference]).
- 1 September 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 73 (5) , 2761-2781
Convolution Bounds on Quantile Aggregation
Advancing Risk Assessment: New Ways To Compute Quantile Aggregation This issue features a pivotal study on quantile aggregation amid dependence uncertainty, an area critical to finance, risk management, and statistics. The authors introduce “convolution bounds,” derived from a recent inf-convolution formula of quantiles and related risk measures. The obtained analytical tools unify existing results and enhance the understanding of quantile methods by providing general, sharp, and computationally efficient solutions. The results offer insights into the extremal dependence structures, with several implications in risk management and economic analysis applications. For more detailed insights, read the full paper, “Convolution Bounds on Quantile Aggregation” (reference: [insert reference]).
- 8 April 2025
- Published by Elsevier in Insurance: Mathematics and Economics
- Vol. 123, 103110
Portfolio benchmarks in defined contribution pension plan management
- 8 April 2025
- Published by Elsevier in Insurance: Mathematics and Economics
- Vol. 123, 103110
Portfolio benchmarks in defined contribution pension plan management
- 20 March 2025
- Published by Springer Nature in Finance and Stochastics
- Vol. 29 (2) , 469-518
A framework of state-dependent utility optimisation with general benchmarks
- 20 March 2025
- Published by Springer Nature in Finance and Stochastics
- Vol. 29 (2) , 469-518
A framework of state-dependent utility optimisation with general benchmarks