2004年, 第17卷, 第3期　刊出日期：2004-07-15

  

• 全选
  |

论文
• Select
  STATISTICS OF A CLASS OF MARKOV CHAINS

  Ying Chun DENG;Xian Bing CAO

  Journal of Systems Science and Complexity. 2004, 17(3): 287-398.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  In this paper we prove that the distributions of their sojourn time and hitting time at one special state for random walks which are allowed to be finite or infinite and Markov chains on star-graphs with discrete time can uniquely determine the probability distribution of the whole chains. This result also suggests a new statistical method for Markov chains.
• Select
  A GENERAL APPROACH BASED ON AUTOCORRELATION TO DETERMINE INPUT VARIABLES OF NEURAL NETWOEKS FOR TIME SERIES FORECASTING

  Wei HUANG;Yoshiteru NAKAMORI;Shou Yang WANG

  Journal of Systems Science and Complexity. 2004, 17(3): 297-305.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  Input selection is probably one of the most critical decision issues in neural network designing, because it has a great impact on forecasting performance. Among the many applications of artificial neural networks to finance, time series forecasting is perhaps one of the most challenging issues. Considering the features of neural networks, we propose a general approach called Autocorrelation Criterion (AC) to determine the inputs variables for a neural network. The purpose is to seek optimal lag periods, which are more predictive and less correlated. AC is a date-driven approach in that there is no prior assumption about the models for time series under study. So it has extensive applications and avoids a lengthy experimentation and tinkering in input selection. We apply the approach to the determination of input vatiables for foreign exchange rate forecasting and conduct comparisons between AC and information-based in-sample model selection criterion. The experiment results show that AC outperforms information-based in-sample modek selection criterion.
• Select
  OBSERVABILITY OF EXTENDED TIMED EVENT GRAPH

  Zhi Bing ZHUO;Wen De CHEN

  Journal of Systems Science and Complexity. 2004, 17(3): 306-309.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  In this paper, we study some results of extended timed event graph (ETEG) by using graph theory's methods in the dioid framework. A necessary and sufficient condition for the observability of ETEG is obtained and ETEG's standard structure is also established.
• Select
  OSCILLATION OF THE SOLUTIONS OF IMPULSIVE HYPERBOLIC EQUATIONS WITH DELAY

  Jiao Wan LUO

  Journal of Systems Science and Complexity. 2004, 17(3): 310-317.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  Oscillation properties of the solutions of impulsive hyperbolic equations with delay are investigated via the method of differential inequalities.
• Select
  DISTURBANCE ATTENUATION FOR UNCERTAIN NONLINEAR CASCADED SYSTEMS

  Wei Ping BI;Xiao Wu MU;Yu Qiang SUN

  Journal of Systems Science and Complexity. 2004, 17(3): 318-324.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  In present paper, the disturbance attenuation problem of uncertain nonlinear cascaded systems is studied. Based on the adding one power integrator technique and recursive design, a feedback controller that solves the disturbance attenuation problem is constructed for uncertain nonlinear cascaded systems with internal stability.
• Select
  EXISTENCE AND NONEXISTENCE OF POSITIVE RADIAL SOLUTIONS FOR A CLASS OF SEMILINEAR ELLIPTIC SYSTEMS

  Jin Biao ZHONG;Zu Chi CHEN

  Journal of Systems Science and Complexity. 2004, 17(3): 325-331.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  In this paper an existence theorem of positive radial solutions to a class of semilinear elliptic systems is proved by the Leray-Schauder degree theorem. Also, a nonexistence theorem is obtained. As an application of the main theorem, an example is given.
• Select
  NONMONOTONIC TRUST REGION PROJECTED REDUCED HESSIAN ALGORITHM WITH TWO-PIECE UPDATE FOR CONSTRAINED OPTIMIZATION

  De Tong ZHU

  Journal of Systems Science and Complexity. 2004, 17(3): 332-348.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  This paper proposes a two-piece update of projected reduced Hessian algorithm with nonmonotonic trust region strategy for solving nonlinear equality constrained optimization problems. In order to deal with large problems, a two-piece update of two-side projected reduced Hessian is used to replace full Hessian matrix. By adopting the Fletcher's penalty function as the merit function, a nonmonotonic trust region strategy is suggested which does not require the merit function to reduce its value in every iteration. The two-piece update of projected reduced Hessian algorithm which switches to nonmonotonic trust region technique possesses global convergence while maintaining a two-step Q-superlinear local convergence rate under some reasonable conditions. Furthermore, one step Q-superlinear local convergence rate can be obtained if at least one of the update formulas is updated at each iteration by an alternative update rule. The numerical experiment results are reported to show the effectiveness of the proposed algorithm.
• Select
  SELF-ADAPTIVE CONTROLS OF A COMPLEX CELLULAR SIGNALING TRANSDUCTION SYSTEM

  Hong LI;Zhi Yuan ZHOU;Rong Yang DAI;Bo LUO;Xiao Li ZHENG;Wen Li YANG;Tao HE;Ming Lu WU

  Journal of Systems Science and Complexity. 2004, 17(3): 349-368.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  In cells, the interactions of distinct signaling transduction pathways originating from cross-talkings between signaling molecules give rise to the formation of signaling transduction networks, which contributes to the changes (emergency) of kinetic behaviors of signaling system compared with single molecule or pathway. Depending on the known experimental data, we have constructed a model for complex cellular signaling transduction system, which is derived from signaling transduction of epidermal growth factor receptor in neuron. By the computational simulating methods, the self-adaptive controls of this system have been investigated. We find that this model exhibits a relatively stable self-adaptive system, especially to over-stimulation of agonist, and the amplitude and duration of signaling intermediates in it could be controlled by multiple self-adaptive effects, such as ``signal scattering", ``positive feedback", ``negative feedback" and ``B-Raf shunt". Our results provide an approach to understanding the dynamic behaviors of complex biological systems.
• Select
  ANALYSIS OF THE MECHANISM MODELS OF TECHNOLOGICAL INNOVATION DIFFUSION

  Jiu Ping XU;Mi Nan HU

  Journal of Systems Science and Complexity. 2004, 17(3): 369-376.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  This paper analyzes the mechanism and principle of diffusion of technology diffusion on the basis of quantitative analysis. Then it sets up the diffusion model of innovation incorporating price, advertising and distribution, the diffusion model of innovation including various kinds of consumers, and the substitute model between the new technology and the old one applied systems dynamics, optimization method, probabilistic method and simulation method on computer. Finally this paper concludes with some practical observations from a case study.
• Select
  CLOSURE AXIOMS FOR POSET MATROIDS

  Shu Chao LI;Yan Qin FENG

  Journal of Systems Science and Complexity. 2004, 17(3): 377-386.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  An excellent introduction to the topic of poset matroids is due to M. Barnabei, G. Nicoletti and L. Pezzoli. In this paper, we extensively study the {closure operators} of poset matroids and obtain the closure axioms for poset matroids; thereby we can characterize poset matroids in terms of the closure axioms. Some corresponding properties of combinatorial schemes are also obtained.
• Select
  ROBUST \mbox{\boldmath$H_\infty$} FILTERING FOR UNCERTAIN DISCRETE-TIME STATE-DELAYED SYSTEMS: A PARAMETER-DEPENDENT LYAPUNOV APPROACH

  Hui Jun GAO;Chang Hong WANG

  Journal of Systems Science and Complexity. 2004, 17(3): 399-411.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  This paper is concerned with the problem of robust $H_\infty$ filtering for linear discrete-time systems with multiple state delays and polytopic uncertain parameters. Attention is focused on the design of full-order, reduced-order and zeroth-order robust $H_\infty$ filters on the basis of a recently published parameter-dependent Lyapunov stability result. Sufficient conditions for the existence of such filters are formulated in terms of linear matrix inequalities, upon which admissible filters can be obtained from convex optimization problems. The proposed methodology has been shown, via a numerical example, to be much less conservative than previous filter design methods in the quadratic framework.
• Select
  GLOBAL CONVERGENCE PROPERTIES OF THREE-TERM CONJUGATE GRADIENT METHOD WITH NEW-TYPE LINE SEARCH

  Chang Yu WANG;Shou Qiang DU;Yuan Yuan CHEN

  Journal of Systems Science and Complexity. 2004, 17(3): 412-420.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  In this paper, a new Wolfe-type line search and a new Armijo-type line search are proposed, and some global convergence properties of a three-term conjugate gradient method with the two line searches are proved.
• Select
  CONVEXIFICATION AND CONCAVIFICATION METHODS FOR SOME GLOBAL OPTIMIZATION PROBLEMS

  Zhi You WU;Lian Sheng ZHANG;Fu Sheng BAI;Xin Min YANG

  Journal of Systems Science and Complexity. 2004, 17(3): 421-436.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  In this paper, firstly, we propose several convexification and concavification transformations to convert a strictly monotone function into a convex or concave function, then we propose several convexification and concavification transformations to convert a non-convex and non-concave objective function into a convex or concave function in the programming problems with convex or concave constraint functions, and propose several convexification and concavification transformations to convert a non-monotone objective function into a convex or concave function in some programming problems with strictly monotone constraint functions. Finally, we prove that the original programming problem can be converted into an equivalent concave minimization problem, or reverse convex programming problem or canonical D.C. programming problem. Then the global optimal solution of the original problem can be obtained by solving the converted concave minimization problem, or reverse convex programming problem or canonical D.C. programming problem using the existing algorithms about them.
• Select
  AN ALGORITHM FOR MAXIMUM LIKELIHOOD ESTIMATES TO SEMI-PARAMETRIC REGRESSION MODEL WITH INTERVAL CENSORED DATA

  Ai Jun ZHANG;Jia Ding CHEN;Li Ping LIU;Qing Cheng WANG

  Journal of Systems Science and Complexity. 2004, 17(3): 437-444.

  摘要 ( ) PDF全文 ( )   可视化   收藏

  In this paper we study the practical procedure for getting the maximum likelihood estimates in a semi-parametric regression model with interval censored data. On the basis of the on previous theoretical results, we give the detailed algorithms when there are one or two covariates in the model.