A5066155007- Risk and Portfolio Optimization
- Financial Risk and Volatility Modeling
- Probability and Risk Models
- Insurance and Financial Risk Management
- Statistical Distribution Estimation and Applications
- Insurance, Mortality, Demography, Risk Management
- Stochastic processes and financial applications
- Decision-Making and Behavioral Economics
- Fuzzy Systems and Optimization
- Statistical Methods and Inference
- Probabilistic and Robust Engineering Design
- Multi-Criteria Decision Making
- Economic theories and models
- Market Dynamics and Volatility
- Economic and Environmental Valuation
- Agricultural risk and resilience
- Mathematical functions and polynomials
- Health Systems, Economic Evaluations, Quality of Life
- Bayesian Methods and Mixture Models
- Risk and Safety Analysis
- RNA modifications and cancer
- Mathematical Inequalities and Applications
- Statistical Methods and Bayesian Inference
- Risk Management in Financial Firms
- Monetary Policy and Economic Impact
University of Science and Technology of China
2015-2024
Shanghai Tenth People's Hospital
2023-2024
Tongji University
2023-2024
Ningbo University
2020
University of Waterloo
2014-2015
Second Affiliated Hospital of Zhengzhou University
2013
We incorporate a notion of risk aversion favoring prudent decisions from financial institutions into regulatory capital calculation principles. In the context Basel III and IV as well Solvency II, is carried out through tools monetary measures. The that we focus on has four equivalent formulations: consistency with second-order stochastic dominance, conditional expectations, or portfolio diversification, expected social impact. class measures representing this referred to consistent...
Inspired by the recent developments in risk sharing problems for value at (VaR), expected shortfall (ES), and range (RVaR), we study optimization of general tail measures. Explicit formulas inf-convolution Pareto-optimal allocations are obtained case a mixed collection left right VaRs, that VaR another measure. The measures is shown to be measure with an aggregated parameter, phenomenon very similar cases VaR, ES, RVaR. Optimal settings elliptical models model uncertainty. In particular,...
Optimal Decision Making Under Distorted Expectation with Partial Distribution Information makers who are not risk neutral may evaluate expected values by distorting objective probabilities to reflect their attitudes, a phenomenon known as distorted expectations. This concept is widely applied in behavioral economics, insurance, finance, and other business domains. In “Distributionally Robust Optimization Expectations,” Cai, Li, Mao study how decision using expectations can optimize decisions...
The purpose of this article is to present several equivalent characterizations comparing the largest-order statistics and sample ranges two sets n independent exponential random variables with respect different stochastic orders, where in one set are heterogeneous other identically distributed. main results complement extend known literature. geometric distribution can be regarded as discrete counterpart distribution. We also study orderings from point out similarities differences between variables.
Abstract Motivated by recent advances on elicitability of risk measures and practical considerations optimization, we introduce the notions Bayes pairs measures. are counterpart elicitable measures, extensively studied in literature. The Expected Shortfall (ES) is most important coherent measure both industry practice academic research finance, insurance, management, engineering. One our central results that under a continuity condition, ES only class We further show entropic which Bayes....
Abstract Expectiles have received increasing attention as a risk measure in management because of their coherency and elicitability at the level $\alpha\geq1/2$ . With view to practical assessments, this paper delves into worst-case expectile, where only partial information on underlying distribution is available there no closed-form representation. We explore asymptotic behavior expectile two specified ambiguity sets: one through Wasserstein distance from reference transforms problem convex...
Generalized quantiles of a random variable were defined as the minimizers general asymmetric loss function, which include quantiles, expectiles and M -quantiles their special cases. Expectiles have been suggested potentially better alternatives to both Value-at-Risk expected shortfall risk measures. In this paper, we first establish first-order expansions generalized for extreme risks confidence level α↑ 1, then investigate and/or second-order an 1 according underlying distribution belonging...
The purpose of this study is two-fold. First, we investigate further properties the second-order regular variation (2RV). These include preservation 2RV under composition operation and generalized inverse transform, among others. Second, derive expansions tail probabilities convolutions non-independent identically distributed (i.i.d.) heavy-tail random variables, establish risk concentration mild assumptions. main results extend some ones in literature from i.i.d. case to non-i.i.d. case.
Abstract We study a distributionally robust reinsurance problem with the risk measure being an expectile and under expected value premium principle. The mean variance of ground-up loss are known, but distribution is otherwise unspecified. A minimax formulated its inner maximization over all distributions known variance. show that equivalent to maximizing three-point distributions, reducing infinite-dimensional optimization finite-dimensional problem. can be solved numerically. Numerical...
A basic assumption of the classic reinsurance model is that distribution loss precisely known. In practice, only partial information available for due to lack data and estimation error. We study a distributionally robust problem by minimizing maximum Value-at-Risk (or worst-case VaR) total retained insurer all distributions with known mean variance. Our handles typical stop-loss contracts. show three-point achieves VaR insurer, from which closed-form solutions optimal deductible are...