Xiaobai Zhu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xiaobai Zhu has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Jinggong Zhang
Nanyang Technological University
3 shared publications - Kenneth Q. Zhou
University of Waterloo
3 shared publications - Yiying Zhang
Southern University of Science and Technology
3 shared publications - Danping Li
East China Normal University
2 shared publications Lv Chen
2 shared publicationsYaodi Yong
2 shared publications
Latest Publications
- 13 November 2026
- Published by Elsevier in Insurance: Mathematics and Economics
Interplay of prevention efforts and insurance demand with distortion risk measures
- 13 November 2026
- Published by Elsevier in Insurance: Mathematics and Economics
Interplay of prevention efforts and insurance demand with distortion risk measures
- Open Access
- 15 October 2026
- Published by Elsevier in SSRN Electronic Journal
Optimal Transition Schedules for Retirement Age Increases: Welfare and Intergenerational Fairness
In response to demographic shifts such as declining fertility rates and increasing life expectancy, many countries are raising the full retirement age (FRA). This paper develops a novel optimization framework for designing FRA increase schedules, explicitly balancing the minimization of aggregate welfare loss across affected generations with intergenerational fairness considerations. Within a continuous-time framework, we model welfare loss through disutility functions applied to actuarial losses generated by pension reforms, while incorporating smoothness penalties on the reform schedule to prevent abrupt intergenerational changes. We derive analytical characterizations of the optimal schedules under linear, exponential, and power disutility, with closed-form expressions available in certain special cases. Moreover, we construct efficient frontiers that visualize the inherent trade-off between welfare loss and intergenerational inequality. Robustness checks using alternative inequality measures, alternative disutility specifications, and binding deficit recovery constraints confirm the qualitative validity of our findings. The deficit-constrained case yields a novel triphasic reform structure, with sharp increases for older workers, a slowdown for middle-aged cohorts, and renewed acceleration for the youngest, which aligns with the structure observed in the U.S. Social Security reform of 1983. Our framework provides policymakers with a transparent toolkit to evaluate the critical trade-offs between efficiency and fairness.
- 15 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessOptimal Transition Schedules for Retirement Age Increases: Welfare and Intergenerational Fairness
In response to demographic shifts such as declining fertility rates and increasing life expectancy, many countries are raising the full retirement age (FRA). This paper develops a novel optimization framework for designing FRA increase schedules, explicitly balancing the minimization of aggregate welfare loss across affected generations with intergenerational fairness considerations. Within a continuous-time framework, we model welfare loss through disutility functions applied to actuarial losses generated by pension reforms, while incorporating smoothness penalties on the reform schedule to prevent abrupt intergenerational changes. We derive analytical characterizations of the optimal schedules under linear, exponential, and power disutility, with closed-form expressions available in certain special cases. Moreover, we construct efficient frontiers that visualize the inherent trade-off between welfare loss and intergenerational inequality. Robustness checks using alternative inequality measures, alternative disutility specifications, and binding deficit recovery constraints confirm the qualitative validity of our findings. The deficit-constrained case yields a novel triphasic reform structure, with sharp increases for older workers, a slowdown for middle-aged cohorts, and renewed acceleration for the youngest, which aligns with the structure observed in the U.S. Social Security reform of 1983. Our framework provides policymakers with a transparent toolkit to evaluate the critical trade-offs between efficiency and fairness.
- Open Access
- 27 August 2026
- Published by Elsevier in SSRN Electronic Journal
Risk sharing with government protection under distorted probabilities
This paper studies a three-layer risk-sharing arrangement among the policyholder, the government and the insurer. The government provides the first-layer limited compensation protection, the insurer covers excess losses subject to a VaR-based indemnity constraint, and the policyholder retains any residual risk. The policyholder decides the optimal coverage limit from the government-sponsored protection and the optimal indemnity from private insurance in order to minimize total risk exposure measured via distorted probabilities. Explicit solutions are derived under a general distortion function and a convex government surcharge function. Robustness and extensions are also discussed, including the case of a concave surcharge function, and we formulate two Stackelberg games to study i) the optimal safety loading for a profit-maximizing insurer, and ii) the optimal surcharge rate set by the government. Numerical analyses are provided to visualize the theoretical results and to illustrate how optimal strategies adapt to the surcharge structure, the insurer's risk-bearing capacity, risk attitude, and premium loading.
- 27 August 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessRisk sharing with government protection under distorted probabilities
This paper studies a three-layer risk-sharing arrangement among the policyholder, the government and the insurer. The government provides the first-layer limited compensation protection, the insurer covers excess losses subject to a VaR-based indemnity constraint, and the policyholder retains any residual risk. The policyholder decides the optimal coverage limit from the government-sponsored protection and the optimal indemnity from private insurance in order to minimize total risk exposure measured via distorted probabilities. Explicit solutions are derived under a general distortion function and a convex government surcharge function. Robustness and extensions are also discussed, including the case of a concave surcharge function, and we formulate two Stackelberg games to study i) the optimal safety loading for a profit-maximizing insurer, and ii) the optimal surcharge rate set by the government. Numerical analyses are provided to visualize the theoretical results and to illustrate how optimal strategies adapt to the surcharge structure, the insurer's risk-bearing capacity, risk attitude, and premium loading.
- Open Access
- 10 March 2026
- Published by Elsevier in SSRN Electronic Journal
A National-Provincial Stochastic Mortality Model for China: Methods, Innovations, and Policy Applications
China's rapid population aging creates urgent demand for reliable mortality forecasts to support pension, healthcare, and insurance planning. This paper develops a national-provincial stochastic mortality model that jointly projects mortality for China and its provinces. The model introduces three main innovations: a structural dependence linking national and provincial mortality through a relative Lee-Carter framework; a Poisson-Gaussian error structure with imputation methods to fully utilize sparse and uneven provincial data; and a flexible provincial trend specification that allows either mean-reverting or unit-root dynamics, selected using the Expected Deviance Information Criterion (EDIC) rather than imposing strict long-run coherence. Estimated using a Bayesian estimation procedure that mitigates key limitations in the data, the model shows strong empirical performance and supports several applications, including quantifying provincial inequality in China's pension annuity factors.
- 10 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessA National-Provincial Stochastic Mortality Model for China: Methods, Innovations, and Policy Applications
China's rapid population aging creates urgent demand for reliable mortality forecasts to support pension, healthcare, and insurance planning. This paper develops a national-provincial stochastic mortality model that jointly projects mortality for China and its provinces. The model introduces three main innovations: a structural dependence linking national and provincial mortality through a relative Lee-Carter framework; a Poisson-Gaussian error structure with imputation methods to fully utilize sparse and uneven provincial data; and a flexible provincial trend specification that allows either mean-reverting or unit-root dynamics, selected using the Expected Deviance Information Criterion (EDIC) rather than imposing strict long-run coherence. Estimated using a Bayesian estimation procedure that mitigates key limitations in the data, the model shows strong empirical performance and supports several applications, including quantifying provincial inequality in China's pension annuity factors.
- 21 November 2025
- Published by Elsevier in Insurance: Mathematics and Economics
- Vol. 126
Welfare-enhancing annuity divisor for notional defined contribution design
- 21 November 2025
- Published by Elsevier in Insurance: Mathematics and Economics
- Vol. 126
Welfare-enhancing annuity divisor for notional defined contribution design
- Open Access
- 20 November 2025
- Published by Elsevier in Insurance: Mathematics and Economics
- Vol. 126
Mortality modeling via vitality: Model constructions and actuarial applications
- 20 November 2025
- Published by Elsevier in Insurance: Mathematics and Economics
- Vol. 126
Open AccessMortality modeling via vitality: Model constructions and actuarial applications
- 13 August 2025
- Published by Elsevier in European Journal of Operational Research
- Vol. 330 (1) , 199-216
Self-protection and insurance demand under time-inconsistent mean–variance framework
- 13 August 2025
- Published by Elsevier in European Journal of Operational Research
- Vol. 330 (1) , 199-216
Self-protection and insurance demand under time-inconsistent mean–variance framework
- 4 April 2025
- Published by Elsevier in Insurance: Mathematics and Economics
- Vol. 122, 275-299
Equilibrium intergenerational risk-sharing design for a target benefit pension plan
- 4 April 2025
- Published by Elsevier in Insurance: Mathematics and Economics
- Vol. 122, 275-299
Equilibrium intergenerational risk-sharing design for a target benefit pension plan
- Open Access
- 21 March 2025
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
- Vol. 189 (2) , 973-997
A Bayesian generalized additive model approach for forecasting mortality improvement with external information
Mortality modelling is facing new challenges as historical mortality experiences are insufficient to foresee unprecedented changes, such as the impact of the COVID-19 pandemic. Expert opinion has become one important source of information that provides additional insights into the pandemic’s possible future courses. In this paper, we develop a Bayesian generalized additive model where external information can be seamlessly integrated into the projection of future mortality improvement (MI) rates. A collection of spline functions over the age and period dimensions is utilized to construct a smooth transition of MI trends from recent changes to long-term rates. Our modelling approach is able to incorporate different types of external information and elicit expert opinions in a coherent probabilistic manner. Lastly, we use three case studies with COVID-19 mortality data to illustrate the applications of the proposed model in different modelling scenarios.
- 21 March 2025
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
- Vol. 189 (2) , 973-997
Open AccessA Bayesian generalized additive model approach for forecasting mortality improvement with external information
Mortality modelling is facing new challenges as historical mortality experiences are insufficient to foresee unprecedented changes, such as the impact of the COVID-19 pandemic. Expert opinion has become one important source of information that provides additional insights into the pandemic’s possible future courses. In this paper, we develop a Bayesian generalized additive model where external information can be seamlessly integrated into the projection of future mortality improvement (MI) rates. A collection of spline functions over the age and period dimensions is utilized to construct a smooth transition of MI trends from recent changes to long-term rates. Our modelling approach is able to incorporate different types of external information and elicit expert opinions in a coherent probabilistic manner. Lastly, we use three case studies with COVID-19 mortality data to illustrate the applications of the proposed model in different modelling scenarios.
- 28 August 2024
- Published by Taylor & Francis in North American Actuarial Journal
- Vol. 29 (1) , 119-143
Optimal Design of Raising Retirement Age
- 28 August 2024
- Published by Taylor & Francis in North American Actuarial Journal
- Vol. 29 (1) , 119-143
Optimal Design of Raising Retirement Age