On representation and regularity of continuous parameter multivalued martingales

In this paper we study multivalued martingales in continuous time. First we show that every multivalued martingale in continuous time can be represented as the closure of a sequence of martingale selections. Then we prove two results concerning the cadlag modifications of continuous time multivalued martingales, in Kuratowski-Mosco convergence and in convergence in the Hausdorff metric respectively.

     

Bootstrapping Autoregressive and Moving Average Parameter Estimates of Infinite Order Vector Autoregressive Processes

We consider anr-dimensional multivariate time series {y_t, tZ} which is generated by an infinite order vector autoregressive process. We show that a bootstrap procedure which works by generating time series replicates via an estimated finitek-order vector autoregressive process (k→∞ at an appropriate rate with the sample size) gives asymptotically valid approximations to the joint distribution of the growing set of estimated autoregressive coefficients and to the corresponding set of estimated moving average coefficients (impuls responses).     

NEW METHOD TO ESTIMATE SCALING OF POWER-LAW DEGREE DISTRIBUTION AND HIERARCHICAL NETWORKS

A new method and corresponding numerical procedure are introduced to estimate scaling exponents of power-law degree distribution and hierarchical clustering func tion for complex networks. This method can overcome the biased and inaccurate faults of graphical linear fitting methods commonly used in current network research. Furthermore, it is verified to have higher goodness-of-fit than graphical methods by comparing the KS (Kolmogorov-Smirnov) test statistics for 10 CNN (Connecting Nearest-Neighbor)networks.     

Sorption phenomena of organic solvents in polymers: Part I

In this work we use the vapor-sorption equilibrium data to show the degree of solvent upturn in each solvent-polymer system. For this purpose, 23 isothermal data sets for four polymer + solvent binaries, one block copolymer + solvent binary and for the corresponding polymer pairs have been used in the temperature range of 25-70 °C. Solvents studied are benzene, carbon tetrachloride, chloroform and pentane. Homopolymers studied are polyisobutylene, poly(ε-caprolactone), poly(ethylene oxide), n-heptadecane, polystyrene, poly(vinyl chloride), poly(vinyl methyl ether), and n-tetracosane.According to these data sets, solvent weight fraction in the polymer is plotted against solvent-vapor activity that is calculated assuming an ideal gas phase of pure solvent vapor neglecting the vapor pressure of the polymer. We use the Flory-Huggins theory to obtain dimensionless interaction parameter, χ. Also the Zimm-Lundberg clustering theory and non-ideality thermodynamic factor, Γ are used to interpret the equilibrium data.     

Regularized collocation method for Fredholm integral equations of the first kind

In this paper we consider a collocation method for solving Fredholm integral equations of the first kind, which is known to be an ill-posed problem. An “unregularized” use of this method can give reliable results in the case when the rate at which smallest singular values of the collocation matrices decrease is known a priori. In this case the number of collocation points plays the role of a regularization parameter. If the a priori information mentioned above is not available, then a combination of collocation with Tikhonov regularization can be the method of choice. We analyze such regularized collocation in a rather general setting, when a solution smoothness is given as a source condition with an operator monotone index function. This setting covers all types of smoothness studied so far in the theory of Tikhonov regularization. One more issue discussed in this paper is an a posteriori choice of the regularization parameter, which allows us to reach an optimal order of accuracy for deterministic noise model without any knowledge of solution smoothness.     

Portfolio selection under strict uncertainty: A multi-criteria methodology and its application to the Frankfurt and Vienna Stock Exchanges

In modern portfolio theory, it is common practice to first compute the risk-reward efficient frontier and then to support an individual investor in selecting a portfolio that meets his/her preferences for profitability and risk. Potential flaws include (a) the assumption that past data provide sufficient evidence for predicting the future performances of the securities under consideration and (b) the necessity to mathematically determine or approximate the investor’s utility function. In this paper, we propose a methodology whose initial phase filters portfolios that are inefficient from a historical perspective. While this is consistent with traditional approaches, the second phase differs from the standard approach as it uses a decision table constructed by considering multiple scenarios assuming strict uncertainty. The table cells measure consequences by a multi-criteria linear performance index of simulated future returns, which avoids difficulties with performance ratios. The real world applicability is illustrated through two studies based on data from the stock exchanges in Frankfurt and Vienna.     

Nonlinear optimization with GAMS /LGO

The Lipschitz Global Optimizer (LGO) software integrates global and local scope search methods, to handle a very general class of nonlinear optimization models. Here we discuss the LGO implementation linked to the General Algebraic Modeling System (GAMS). First we review the key features and basic usage of the GAMS /LGO solver option, then present reproducible numerical results to illustrate its performance.     

局部形状可调整的三次有理B样条插值曲线

利用三次非均匀有理B样条,给出了一种构造局部插值曲线的方法,生成的插值曲线是C2连续的.曲线表示式中带有一个局部形状参数,随着一个局部形状参数值的增大,所给曲线将局部地接近插值点构成的控制多边形.基于三次非均匀有理B样条函数的局部单调性和一种保单调性的准则,给出了所给插值曲线的保单调性的条件.     

非寿险分类费率模型及其参数估计

在非寿险分类费率厘定中,存在各种模型可供选择,如加法模型、乘法模型、混合模型和广义线性模型等,而在这些模型的参数估计中,还存在各种可供选择的估计方法,如最小二乘法、极大似然法、最小x2法、直接法和边际总和法等。这些模型和参数估计方法散见于各种精算学文献中,本文对这些模型和参数估计方法进行了系统的比较和分析,并揭示了它们之间存在的一些等价关系。     

最大顶点覆盖问题的一种近似算法

给出了求解最大顶点覆盖问题的一种近似算法,讨论了它的性能保证,利用P ipage技术,为最大顶点覆盖问题设计出了0.75-近似算法.