Loading...
Association Journal of CSIAM
Supervised by Ministry of Education of PRC
Sponsored by Xi'an Jiaotong University
ISSN 1005-3085  CN 61-1269/O1

Table of Content

    15 February 2022, Volume 39 Issue 1 Previous Issue    Next Issue
    Optimal Reinsurance and Investment Strategies for Insurers with Ambiguity Aversion: Minimizing the Probability of Ruin
    WANG Yuwei, RONG Ximin, ZHAO Hui
    2022, 39 (1):  1-19.  doi: 10.3969/j.issn.1005-3085.2022.01.001
    Abstract ( 42 )   PDF (15913KB) ( 35 )   Save
    The paper studies an optimal investment and reinsurance problem for an ambiguity-averse insurer (AAI) who aims to minimize the ruin probability under model ambiguity. The insurer is allowed to purchase a proportional reinsurance and invest in one risky asset. The surplus process of the insurer is described by a diffusion risk model and the price process of risky asset is described by the constant elasticity variance (CEV) model. According to the dynamic programming principle, the paper derives the corresponding Hamilton-Jacobi-Bellman (HJB) equation. The optimal strategy and value function are obtained explicitly for special elasticity coefficients. Finally, numerical models illustrate the effects of model parameters on optimal strategy and value function. The paper finds that, the investment and reinsurance strategy of the insurer becomes more conservative when the insurer is more ambiguity averse.
    Related Articles | Metrics
    Optimal Reinsurance Strategy Game between Insurer and Reinsurer for a Diffusion Risk Model
    LIN Xiang, ZHU Guanxia, QIAN Yiping
    2022, 39 (1):  20-36.  doi: 10.3969/j.issn.1005-3085.2022.01.002
    Abstract ( 41 )   PDF (564KB) ( 33 )   Save
    The insurance company can decide whether to purchase a proportional reinsurance according to the reinsurance price, while the reinsurance company can decide whether to accept the reinsurance contract. Then a reinsurance contract should appear as a reasonable compromise between the insurer's and the reinsurer's interest. The paper considers a reinsurance selection game problem between an insurance company and a reinsurance company for a diffusion risk model by invoking the use of the dynamic programming principle. A closed-form solutions of the reinsurance selection game problems is obtained under three cases for the insurance company and the reinsurance company with an exponential utility function. The expected surplus utilities of the insurance company and the reinsurance company can increase under some conditions. Numerical examples are also provided to illustrate how the reinsurance strategy and the reinsurance price, as well as the utility profit change when some model parameters vary.
    Related Articles | Metrics
    Research on Data Compressed Sensing Algorithm for Spacecraft Structural Health Monitoring
    LI Yu, LI Chen, WANG Changlong, MEI Zhandong, ZHANG Shiyi
    2022, 39 (1):  37-49.  doi: 10.3969/j.issn.1005-3085.2022.01.003
    Abstract ( 41 )   PDF (2776KB) ( 21 )   Save
    The structural health monitoring of a spacecraft product is an important process to ensure its safe and stable operation during launch and in orbit. Since multiple sensors will generate a large amount of data during long-term monitoring and require efficient transmission and storage, this article is aimed at data transmission problem puts forward and proves a fractional minimization model algorithm based on sparse recovery technology and an improved fractional screening algorithm, combined with satellite vibration test data and classic matching tracking algorithms for comparative analysis and verification. The comparison results show that, under the environmental conditions of the spacecraft, the relative error of data recovery obtained by the algorithm in this paper is lower than that of the two existing algorithms. The accuracy of data recovery is guaranteed under the condition of higher compression ratio, and hence we implement the data recovery from low-dimensional data to high-dimensional data transmission.
    Related Articles | Metrics
    Structure Learning of Directed Acyclic Graphs Incorporating the Scale-free Prior
    SU Wenqing, GUO Xiao, ZHANG Hai
    2022, 39 (1):  50-62.  doi: 10.3969/j.issn.1005-3085.2022.01.004
    Abstract ( 51 )   PDF (693KB) ( 29 )   Save
    Graphical model is an effective method to analyze the network structure, in which directed acyclic graphs have been widely used to model the causal relationships among variables. While many real networks are scale-free, that is, the degree of the network follows a power-law. The paper considers the problem of structure learning in directed acyclic graphs incorporating the scale-free prior. Specifically, we assume the order of nodes is known in advance. To capture the scale-free property, we propose a novel regularization model with a penalty which is the composite of the Log-type and $l_q (0<q<1)$-type penalty functions to solve the non-convex model and to analyze the convergence of the algorithm. Experiments show that the proposed method performs well for both the simulation study and real data applications. 
    Related Articles | Metrics
    Steady-state Analysis and Optimal Production Strategy of Production Service Inventory Model with Server's Vacation
    XU Hao, YUE Dequan
    2022, 39 (1):  63-78.  doi: 10.3969/j.issn.1005-3085.2022.01.005
    Abstract ( 42 )   PDF (219KB) ( 23 )   Save
    The production service inventory system with server's vacation by combining the server's vacation and the production-service inventory is examined in the paper. Demands (customers) arrive according to a Poisson process, and the service time of each customer and the production time of each product are all assumed to be exponentially distributed. When the inventory is depleted, the server goes to a random time of a vacation. It is assumed that the vacation time follows an exponential distribution. Firstly, the steady state condition of the system is obtained by using the quasi-birth-and-death process. Then, the steady-state analysis is established for the case of the system model with negligible service time and the explicit expressions of the steady state probabilities is obtained of this special case. Furthermore, it is proved that the steady-state probability distribution of the system has a structure of product form solution. On this basis, some performance indexes and cost function are calculated. Finally, the optimal $( {s, S})$ inventory strategy of the model is numerically obtained. The numerical analysis is performed to study the influence of some system parameters on performance indexes, the optimal inventory strategy and the optimal cost.
    Related Articles | Metrics
    A Nonstandard Numerical Methods for a Mathematical Model for Cholera with Multiply Delays and Dispersal
    LIAO Shu, FANG Zhangying, YANG Weiming
    2022, 39 (1):  79-92.  doi: 10.3969/j.issn.1005-3085.2022.01.006
    Abstract ( 51 )   PDF (1324KB) ( 32 )   Save
    A spatial and delayed epidemic model with media coverage is studied, and the global asymptotically stability of the equilibria is proved. Furthermore, a discreted edition of this model is established by applying a nonstandard finite difference scheme which has the same dynamics as the original system. Moveover, it is shown that the global stability of the disease-free equilibrium, and the globally asymptotically stability of the endemic equilibrium is studied by constructing a suitable Lyapunov function. Finally, the NSFD scheme can be well simulated to support the theoretical results.
    Related Articles | Metrics
    A Positivity-preserving Lagrangian Method for Ideal Magnetohydrodynamics Equations in One-dimension
    ZOU Shijun, YU Xijun, DAI Zihuan
    2022, 39 (1):  93-106.  doi: 10.3969/j.issn.1005-3085.2022.01.007
    Abstract ( 53 )   PDF (931KB) ( 31 )   Save
    Lagrangian methods play a very important role in computational fluid dynamics and especially suitable for dealing with the physical problems related to high-intensity magnetic field such as Z-pinch, Tokamak, ICF, and so on. In these physical problems the density and thermal pressure are always non-negative. However, such positivity property is not always satisfied by approximated solutions which obtained by a numerical scheme. To deal with this problem, the paper develops a Lagrangian HLLD approximate Riemann solver which can keep positivity-preserving property under some appropriate signal speeds. With this solver, a conservative Lagrangian scheme for solving the ideal compressible magnetohydrodynamics equations in one-dimensional is proposed. At last, some numerical examples are presented to demonstrate the positivity-preserving property of our scheme.
    Related Articles | Metrics
    A Partially Linear Errors-in-Variables Model with Observations at Grid Points
    MA Junling, WANG Lichun
    2022, 39 (1):  107-119.  doi: 10.3969/j.issn.1005-3085.2022.01.008
    Abstract ( 46 )   PDF (189KB) ( 23 )   Save
    For grid data, partially linear errors-in-variables models are constructed and studied. Large sample statistical inferences are made for the models as sample sizes of both two dimensions and one dimension tend to infinity respectively. For the modified least square estimators of the unknown parameters of the linear part and the variance of model error, their strong consistencies and convergence rates are obtained under weak conditions. For the spline function estimator of the unknown function of the nonlinear part, its strong consistency and convergence rate are obtained under certain conditions. Further, for the estimators of the parameters of the linear part and the variance of model error, their asymptotical normality are proved under some regularity conditions as sample sizes of two dimensions and one dimension tend to infinity respectively, and the asymptotical variances are provided.
    Related Articles | Metrics
    A High-order Compact Difference Method for Solving Two-dimensional Variable Coefficients Helmholtz Equation with Discontinuous Wave Number
    WANG Fang, FENG Xiufang
    2022, 39 (1):  120-134.  doi: 10.3969/j.issn.1005-3085.2022.01.009
    Abstract ( 43 )   PDF (202KB) ( 33 )   Save
    Many practical physical problems can be numerically simulated by the variable coefficients Helmholtz equation with discontinuous wave number. The numerical method of the Helmholtz equation, with important theoretical and practical significance, is one of the hot research topics. Due to the discontinuity of wave number, the traditional finite difference method is usually unable to achieve the accuracy of the original difference scheme when solving Helmholtz equation with discontinuous wave number. Based on the idea of immersed interface method, a new fourth-order compact finite difference scheme is constructed  for two-dimensional coefficient Helmholtz equation with discontinuous wave number. The reliability and effectiveness of the new method are verified by numerical experiments.
    Related Articles | Metrics
    Solubility of Boundary Value Problems for Nonlinear Fractional Coupled Functional Differential Systems
    QIAO Ruonan, LIU Xiping, JIA Mei
    2022, 39 (1):  135-147.  doi: 10.3969/j.issn.1005-3085.2022.01.010
    Abstract ( 37 )   PDF (182KB) ( 22 )   Save
    Since the speed of all motions are limited, the time-delay phenomena are often inevitable in the signal transmission or other process. The fractional functional differential equations are important models to study the movement of time-delay systems. When there are two or more interact state variables in the system, they could always be characterized by coupled differential equations. The existence and uniqueness of positive solutions for boundary value problems of a class of nonlinear delay coupled functional differential systems with Riemann-Liouville fractional derivatives are studied. Firstly, according to the characteristics of equations and boundary conditions, a comparison theorem for the system is constructed, the monotonic sequence of upper and lower solutions is obtained, and the relationship between the upper and lower solutions is determined. Secondly, the existence theorems for positive solution of boundary value problem are established and proved by using the method of upper and lower solutions, and the value range of positive solutions is obtained. And thirdly, the existence and uniqueness theorem for positive solution of the boundary value problem is established and proved by iterative technique. Finally, a specific example is given out to illustrate the adaptability and universality of the main results.
    Related Articles | Metrics
    Existence of Solutions to Boundary Value Problem of Multi-term Fractional Differential Equations
    LI Jianli, CHEN Lizhen, LI Gang
    2022, 39 (1):  148-158.  doi: 10.3969/j.issn.1005-3085.2022.01.011
    Abstract ( 39 )   PDF (166KB) ( 35 )   Save
    It is well known that lots of practical problems can be described by the models of fractional differential equations, and the fractional calculus and fractional differential equations are important mathematical tools in dealing the practical problems. By using the contraction mapping principle, Krasnoselskii fixed-point theorem and noncompact measure theory, a class of multi-term fractional differential equations with mixed boundary condition is studied, and some existence and uniqueness results under different conditions to the nonlinear term are obtained. These results generalize the existence theorems for the classical Bagley-Torvik equation and some related models.
    Related Articles | Metrics
    Discharge Behavior Analysis and Control Synchronization of Magnetic Flux e-HR Neuron Model
    MA Jie, GAO Jie, DU Mengmeng, YANG Lixin
    2022, 39 (1):  159-170.  doi: 10.3969/j.issn.1005-3085.2022.01.012
    Abstract ( 38 )   PDF (1786KB) ( 23 )   Save
    Based on the e-HR neuron model, a magnetic flux e-HR neuron model is constructed under the influence of external electromagnetic fields with membrane potential changes. Firstly, the Shengjin formula is used to obtain the number of equilibrium points of the magnetic flux e-HR neuron model and the stability of the equilibrium points, and at the same time the bifurcation theory is further used to find out the conditions for the bifurcation of the system. Secondly, by changing the applied stimulation current and the magnetic flux feedback gain, it is found that the neurons present a variety of firing modes such as cluster firing and periodic firing. What's more, the unstable state of the equilibrium point corresponds to the neuronal firing mode; the steady state corresponds to the resting mode, that is, the non-discharge mode. Finally, an adaptive controller is designed based on the Lyapunov stability theory, and the effectiveness and feasibility of the controller is proved through theoretical analysis and numerical simulation.
    Related Articles | Metrics