基于α-CIS值的非合作-合作双型博弈的求解

本文研究了非合作-合作双型博弈模型求解的问题.首先利用于α-CIS值,求解非合作-合作双型博弈中的合作博弈阶段,再对非合作博弈阶段求其纯策略纳什均衡,获得了基于α-CIS值的双型博弈的一种新的求解方法.推广了原始双型博弈模型的求解方法并证明其可行性.     

具有随机工资的养老金最优投资问题

在三种目标函数下, 研究了具有随机工资的养老金最优投资问题. 第一种是均值-方差准则, 第二种基于效用的随机微分博弈, 第三种基于均值-方差准则的随机微分博弈. 随机微分博弈问题中博弈的双方为养老金计划投资者和金融市场, 金融市场是博弈的虚拟手. 应用线性二次控制理论求得了三种目标函数下的最优策略和值函数的显式解.     

基于注资-阀值分红的随机微分投资-再保博弈

为了更好地反映模型风险对保险公司金融策略的影响,考虑了存在模型风险时,保险公司的最优投资-再保-注资-阀值分红策略问题.在分红与注资总量的贴现值之差的期望最大化的准则下,使用零和随机微分博弈理论建立了保险公司的随机微分博弈模型,通过求解HJBI方程得到了最优投资-再保-注资-阀值分红策略的显式解.最后在有模型风险和无模型风险两种不同情形下,通过数值算例分析了保险公司金融策略之间的差异,为保险资金的管理提供了重要的决策指导.     

考虑时滞效应与均值-方差效用的非零和投资与再保险博弈

在考虑时滞效应的影响下研究了非零和随机微分投资与再保险博弈问题。以最大化终端绝对财富和相对财富的均值-方差效用为目标,构建了两个相互竞争的保险公司之间的非零和投资与再保险博弈模型,分别在经典风险模型和近似扩散风险模型下探讨了博弈的Nash均衡策略。借助随机控制理论以及相应的广义Hamilton-Jacobi-Bellman（HJB）方程,得到了均衡投资与再保险策略和值函数的显式表达。最后,通过数值例子分析了模型中相关参数变动对均衡策略的影响。     

风险资产价格服从CEV模型的投资组合随机微分博弈

本文在风险资产价格服从CEV模型时,讨论两个投资者的时间一致均值-方差最优投资组合选择的随机微分博弈问题.运用动态规划原理,求得了最优投资策略及相应的值函数.     

基于随机微分博弈的最优投资

研究存在模型风险的最优投资决策问题,将该问题刻画为投资者与自然之间的二人-零和随机微分博弈,其中自然是博弈的"虚拟"参与者.利用随机微分博弈分析方法,通过求解最优控制问题对应的HJBI(Hamilton-Jacobi-Bellman-Isaacs)方程,在完备市场和存在随机收益流的非完备市场模型下,都得到了投资者最优投资策略以及最优值函数的解析表达式.结果表明,在完备市场条件下,投资者的最优风险投资额为零,在非完备市场条件下最优投资策略将卖空风险资产,且卖空额随着随机收益流波动率的增大而增加,随风险资产波动率增大而减少.     

BNN动力学下一类连续博弈均衡解的存在性及稳定性

本文研究了一类连续博弈解的存在性及稳定性.利用BNN动力学理论和方法,将演化博弈论中的几个经典例子:鹰-鸽博弈、协调博弈和猜硬币博弈转化为连续型支付函数的连续博弈后,获得了鹰-鸽连续博弈的Nash平衡点是演化稳定和连续稳定的,推广了文献[8]中关于演化博弈Nash平衡点及稳定性结果.     

带加法白噪声的随机三维Camassa-Holm模型的H~2-吸引子

研究带加法白噪声的三维Camassa-Holm模型的随机动力性.通过验证解满足随机Flattening条件,得到随机动力系统的w-极限紧性.然后,利用李和郭关于拟连续随机动力系统的结果,获得该模型的随机动力性的H~2-吸引子描述.     

轨道结构随机场模型与车辆-轨道耦合随机动力分析

将轨道结构视为一个参数随机系统,提出并建立了轨道结构的随机场模型.利用车辆-轨道耦合动力学的基本方法,将轨道系统有限单元模型与多刚体车辆模型相结合,建立了考虑铁路线路参数空-时随机变化的车辆-轨道动力计算模型.算例表明:所提出的方法较为可靠且高效;线路参数随机性对车辆-轨道系统的动力响应有明显的影响,随线路参数离散程度的增加,可能造成行车不安全、轨道损伤加剧等一些问题.     

随机利率下分数跳-扩散Ornstein-Uhlenbeck期权定价模型

假设股票价格遵循分数布朗运动和复合泊松过程驱动的随机微分方程,短期利率服从HullWhite模型,建立了随机利率情形下的分数跳-扩散Ornstein-Uhlenbeck期权定价模型,利用价格过程的实际概率测度和公平保费原理,得到了欧式看涨期权定价的解析表达式,推广了Black-Scholes模型.     

成批到达排队系统的随机比较

本文研究成批到达排队系统中队长过程的随机比较问题．利用随机比较方法我们对成批到达指数服务的多服务台排队系统进行分析,得到了该排队系统中队长过程的随机比较以及队长函数关于时间的凹性和凸性．同时我们也给出了成批到达一般服务的单服务台排队系统队长过程、稳态队长的随机比较以及队长函数关于时间的凹性和凸性．     

A class of stochastic differential equations with pathwise unique solutions

We propose a new method viz., using stochastic partial differential equations to study the pathwise uniqueness of stochastic (ordinary) differential equations. We prove the existence and pathwise uniqueness of a class of stochastic differential equations with coefficients in suitable Hermite-Sobolev class using our approach.     

Full-Discrete Finite Element Method for Stochastic Hyperbolic Equation

This paper is concerned with the finite element method for the stochastic wave equation and the stochastic elastic equation driven by space-time white noise. For simplicity, we rewrite the two types of stochastic hyperbolic equations into a unified form. We convert the stochastic hyperbolic equation into a regularized equation by discretizing the white noise and then consider the full-discrete finite element method for the regularized equation. We derive the modeling error by using "Green's method" and the finite element approximation error by using the error estimates of the deterministic equation. Some numerical examples are presented to verify the theoretical results.     

Variational Inequalities for Combined Control and Stopping Game

Abstract

We study a nonlinear elliptic variational inequality associated with the combined stochastic control problem. By using the dynamic programming principle and the method of penalization, we establish the existence of a unique viscosity solution of the variational inequality and describe it as the value function of the corresponding combined stochastic game problem.     

Potentials Kernels Associated with a Filtration and Forward- Backward SDEs

In this paper, we study some operators associated with a filtration. We prove that these operators are Markov potential kernels closing a sub-Markov resolvent, using the method of continuation inspired by the study of forward-backward stochastic differential equations.     

An Asset Liability Management Model for Casualty Insurers: Complexity Reduction vs. Parameterized Decision Rules

In this paper we study possibilities for complexity reductions in large scale stochastic programming problems with specific reference to the asset liability management (ALM) problem for casualty insurers. We describe a dynamic, stochastic portfolio selection model, within which the casualty insurer maximizes a concave objective function, indicating that the company perceives itself as risk averse. In this context we examine the sensitivity of the solution to the quality and accuracy with which economic uncertainties are represented in the model. We demonstrate a solution method that combines two solution approaches: A truly stochastic, dynamic solution method that requires scenario aggregation, and a solution method based on ex ante decision rules, that allow for a greater number of scenarios. This dynamic/fix mix decision policy, which facilitates a huge number of outcomes, is then compared to a fully dynamic decision policy, requiring fewer outcomes. We present results from solving the model. Basically we find that the insurance company is likely to prefer accurate representation of uncertainties. In order to accomplish this, it will accept to calculate its current portfolio using parameterized decision rules.     

The existence of stationary distribution of a stochastic delayed chemostat model

In this paper, based on the existing literature, we further study an important statistical character of a stochastic delayed chemostat model. By constructing suitable Lyapunov functional and using the stochastic Lyapunov analysis method, we investigate the existence of stationary distribution and the ergodicity of a stochastic delayed chemostat model, which can help us better understand the dynamic behavior and statistical characteristics of stochastic delayed biological models.     

Solvability of forward–backward stochastic partial differential equations

In this paper we study the solvability of a class of fully-coupled forward–backward stochastic partial differential equations (FBSPDEs). These FBSPDEs cannot be put into the framework of stochastic evolution equations in general, and the usual decoupling methods for the Markovian forward–backward SDEs are difficult to apply. We prove the well-posedness of the FBSPDEs, under various conditions on the coefficients, by using either the method of contraction mapping or the method of continuation. These conditions, especially in the higher dimensional case, are novel in the literature.     

Global dynamics analysis of a nonlinear impulsive stochastic chemostat system in a polluted environment

This paper intends to develop a new method to obtain the threshold of an impulsive stochastic chemostat model with saturated growth rate in a polluted environment. By using the theory of impulsive differential equations and stochastic differential equations, we obtain conditions for the extinction and the permanence of the microorganisms of the deterministic chemostat model and the stochastic chemostat model. We develop a new numerical computation method for impulsive stochastic differential system to simulate and illustrate our theoretical conclusions. The biological results show that a small stochastic disturbance can cause the microorganism to die out, that is, a permanent deterministic system can go to extinction under the white noise stochastic disturbance. The theoretical method can also be used to explore the threshold of some impulsive stochastic differential equations.     

Meshfree Approximation for Stochastic Optimal Control Problems

In this work,we study the gradient projection method for solving a class of stochastic control problems by using a mesh free approximation ap-proach to implement spatial dimension approximation.Our main contribu-tion is to extend the existing gradient projection method to moderate high-dimensional space.The moving least square method and the general radial basis function interpolation method are introduced as showcase methods to demonstrate our computational framework,and rigorous numerical analysis is provided to prove the convergence of our meshfree approximation approach.We also present several numerical experiments to validate the theoretical re-sults of our approach and demonstrate the performance meshfree approxima-tion in solving stochastic optimal control problems.