A matrix-based VaR model for risk identification in power supply networks

This paper presents a value-at-risk (VaR) model based on the singular value decomposition (SVD) of a sparsity matrix for voltage risk identification in power supply networks. The matrix-based model provides a more computationally efficient risk assessment method than conventional models such as probability analysis and sensitivity analysis, for example, and provides decision makers in the power supply industry with sufficient information to minimize the risk of network collapse or blackouts. The VaR model is incorporated into a risk identification system (RIS) programmed in the MATLAB environment. The feasibility of the proposed approach is confirmed by performing a series of risk assessment simulations using the standard American Electric Power (AEP) test models (i.e. 14-, 30- and 57-node networks) and a real-world power network (Taiwan power network), respectively. In general, the simulated results confirm the ability of the matrix-based model VaR model to efficient identify risk of power supply networks.     

A generalized linear production model: A unifying model

We introduce a generalized linear production model whose attractive feature being that the resources held by any subset of producersS is not restricted to be the vector sum of the resources held by the members ofS. We provide sufficient conditions for the non-emptiness of the core of the associated generalized linear production game, and show that if the core of the game is not empty then a solution in it can be produced from a dual optimal solution to the associated linear programming problem. Our generalized linear production model is a proper generalization of the linear production model introduced by Owen, and it can be used to analyze cooperative games which cannot be studied in the ordinary linear production model framework. We use the generalized model to show that the cooperative game induced by a network optimization problem in which players are the nodes of the network has a non-empty core. We further employ our model to prove the non-emptiness of the core of two other classes of cooperative games, which were not previously studied in the literature, and we also use our generalized model to provide an alternative proof for the non-emptiness of the core of the class of minimum cost spanning tree games. Thus, it appears that the generalized linear production model is a unifying model which can be used to explain the non-emptiness of the core of cooperative games generated by various, seemingly different, optimization models.This research was partially done while the author was visiting the Graduate School of Business Administration at Tel-Aviv University. The research was partially supported by Natural Sciences and Engineering Research Council Canada Grant A4181 and by SSHRC leave fellowship 451-83-0030.Dedicated to George B. Dantzig.     

A unifying model of strategic network formation

We provide a model that merges two basic models of strategic network formation and incorporates them as extreme cases: Jackson and Wolinsky’s connections model based on bilateral formation of links, and Bala and Goyal’s two-way flow model, where links can be unilaterally formed. In our model a link can be created unilaterally, but when it is only supported by one of the two players the flow through it suffers some friction or decay, but more than when it is supported by both players. When the friction in singly-supported links is maximal (i.e. there is no flow) we have Jackson and Wolinsky’s connections model, while when flow in singly-supported links is as good as in doubly-supported links we have Bala and Goyal’s two-way flow model. In this setting, a joint generalization of the results relative to efficiency and stability in both seminal papers is achieved, and the robustness in both models is tested with positive results.     

A unifying model based on retraction for fixed point algorithms

We present a unifying model based on retraction for several restart fixed point algorithms. The model embraces the interpretation of the algorithms in terms of stationary point problem by van der Laan and Talman and fully explains the 2-ray method.     

G-networks: a unifying model for neural and queueing networks

We survey results concerning a new stochastic network we have developed [1–7], which was initially motivated by neural network modelling [1], or — as we called it — by queueing networks with positive and negative customers [2, 3]. Indeed, it is well known that signals in neural networks are formed by impulses or action potentials, traveling much like customers in a queueing network. We call this model a G-network because it serves as a unifying basis for diverse areas of stochastic modelling in queueing networks, computer networks, computer system performance and neural networks. In its simplest version, negative and positive signals or customers circulate among a finite set of units, modelling inhibitory and excitatory signals of a neural network, or negative and positive customers of a queueing network. Signals can arrive either from other units or from the outside world. Positive signals are accumulated at the input of each unit, and constitute its signal potential. The state of each unit or neuron is its signal potential (which is equivalent to the queue length), while the network state is the vector of signal potentials at each neuron. If its potential is positive, a unit or neuron fires, and sends out signals to the other neurons or to the outside world. As it does so, its signal potential is depleted. In the Markovian case, this model has product form, i.e. the steady-state probability distribution of its potential vector is the product of the marginal probabilities of the potential at each neuron. The signal flow equations of the network, which describe the rate at which positive or negative signals arrive to each neuron, are non-linear. We discuss the relationship between this model and the usual connectionist (formal) model of neural networks, and present applications to combinatorial optimization and to image texture processing. Extensions of the model to the case of multiple signal classes, and to networks with triggered customer motion are presented. We also examine the general stability conditions which guarantee that the network has a well-defined steady-state behaviour.     

A unifying Radon-Nikodym theorem for vector measures

It is a known fact that certain derivation bases from martingales with a directed index set. On the other hand it is also true that the strong convergence of certain abstract martingales is a consequence of the Radon-Nikodym theory for vector measures (cf. Uhl, J. J., Jr., Trans. Amer. Math. Soc.145 1969, 271–285). Many other connections and applications of the latter theory with multidimensional problems in stochastic processes and representation theory are known (cf. Dinculeanu, N., Studia Math.25 1965, 181–205; Dinculeanu, N., and Foias, C., Canad. J. Math.13 1961, 529–556; Rao, M. M., Ann. Mat. pura et applicata76 1967, 107–132; Rybakov, V. I., Izv. Vys?. U?ebn. Zaved. Matematika19 1968, 92–101; Rybakov, V. I., Dokl. Akad. Nauk SSSR180 1968, 620–623). Starting from various vantage points, many authors have proposed several hypotheses for establishing abstract Radon-Nikodym theorems. In view of the great interest and importance of this problem in the areas mentioned above, it is natural to obtain a unifying result with a general enough hypothesis to deduce the various forms of the Radon-Nikodym theorem for vector measures. This should illuminate the Radon-Nikodym theory for vector measures and stimulate further work in abstract martingale problems. In this paper the first problem is attacked, leaving the martingale part and other applications for another study.The main result (Theorem 7 of Section 2) provides the desired unification and from if the Dunford-Pettis theorem, the Phillips theorem and several others are obtained. As martingale-type arguments are constantly present, a careful reader may note the easy translation of the hypothesis to the martingale convergence problem but we treat only the Radon-Nikodym problem using the language of measure theory and linear analysis.     

A unifying framework for teaching probability event types

The concepts and mathematics of mutually exclusive, dependent and independent events are developed in a unifying framework of event association.     

A matrix-based ranking method with application to tennis

We introduce, and investigate, a ranking methodology which may be of interest in sports like tennis. The approach may also be of interest in decision-making situations based on pairwise comparisons. The method is based on linear algebra and one computes a score for each player by solving a certain linear system of equations – from these scores one finds the ranking. The input is a set of matches, and weights representing the importance of the matches; this is represented by a weighted directed graph. We prove a number of properties of the method, including uniqueness of scores, connection to M-matrices and combinatorial interpretations. A case study from ranking in professional tennis is discussed in detail.     

A unifying approach to integration for bounded positive charges

This paper deals with n-monotone functionals, which constitute a generalisation of n-monotone set functions. Using the notion of exactness of a functional, we introduce a new notion of lower and upper integral which subsumes as particular cases most of the approaches to integration in the literature. As a consequence, we can characterise which types of integrals can be used to calculate the natural extension (the lower envelope of all linear extensions) of a positive bounded charge.     

A value for generalized probabilistic communication situations

We introduce in this work an extension of the model of games with probabilistic graphs arising in Calvo et al. (1999, Math. Soc. Sci. 37, 79), which itself generalizes the one developed by Myerson (1977, Math. of Oper. Res. 2, 225) for games with communications restrictions. In the first of these models, each pair of nodes has a given probability of direct communication. In this paper a more general setting is considered: we suppose that a probability distribution over the set of all possible communication networks among the players is given. A generalization of the Myerson value is defined and characterized in this context.     

A matrix-based measure of inter-node walk relatedness in a network

For a pair of nodes in a network, a measure of walk relatedness is introduced. The measure is based on the total weight (number) of k-step walks connecting the pair, i.e., the corresponding entry of the kth power of the network matrix as k → ∞. The damping factor r ^?k is used, where r is the largest eigenvalue of the network matrix. The measure turns out to be equal to the product of the pair’s coreness values, i.e., the nodes’ coordinates in the network matrix’s right and left eigenvectors corresponding to r. The reduction of walk relatedness in a network caused by the removal of a node or link is investigated, i.e., the dependence of the reduction on the structural position (coreness) of the removed element. It is revealed that the “damage” can be measured by the drop in the value of r after the removal; to find this drop, the perturbation method is used. Some possible applications are indicated, and a numerical example with a large real network of 197 nodes and 780 links is considered.     

A unifying theorem on newton's method

In this paper a general theorem is proved about convergence and error estimates for Newton's method. It contains as special cases a theorem of Kantorowitsch-type, a theorem with nonlinear majorants and componentwise error estimates as well as a theorem of monotone type.     

A unifying approach to heuristic search

We present the General Search Procedure (GSP) that provides a unifying way of describing search algorithms. The GSP captures both constructive and iterative search algorithms. We demonstrate as an exercise that various well-known heuristic search procedures can be obtained as instances of the GSP. The introduced formalism provides a solid ground to prove theoretical properties of search methods. Furthermore, by the formal approach we obtain a framework that can serve as the basis of implementing a search based problem solver.     

Active set algorithms for isotonic regression; A unifying framework

In this and subsequent papers we will show that several algorithms for the isotonic regression problem may be viewed as active set methods. The active set approach provides a unifying framework for studying algorithms for isotonic regression, simplifies the exposition of existing algorithms and leads to several new efficient algorithms. We also investigate the computational complexity of several algorithms.In this paper we consider the isotonic regression problem with respect to a complete order where eachw _i is strictly positive and eachy _i is an arbitrary real number. We show that the Pool Adjacent Violators algorithm (due to Ayer et al., 1955; Miles, 1959; Kruskal, 1964), is a dual feasible active set method and that the Minimum Lower Set algorithm (due to Brunk et al., 1957) is a primal feasible active set method of computational complexity O(n ²). We present a new O(n) primal feasible active set algorithm. Finally we discuss Van Eeden's method and show that it is of worst-case exponential time complexity.This work was supported by the National Science and Engineering Research Council of Canada under Research Grant A8189 and an Ontario Graduate Scholarship.     

A composite run-to-the-bank rule for multi-issue allocation situations

In this paper, we propose a new extension of the run-to-the-bank rule for bankruptcy situations to the class of multi-issue allocation situations. We show that this rule always yields a core element and that it satisfies self-duality. We characterise our rule by means of a new consistency property, issue-consistency.     

Context tree selection: A unifying view

Context tree models have been introduced by Rissanen in [25] as a parsimonious generalization of Markov models. Since then, they have been widely used in applied probability and statistics. The present paper investigates non-asymptotic properties of two popular procedures of context tree estimation: Rissanen’s algorithm Context and penalized maximum likelihood. First showing how they are related, we prove finite horizon bounds for the probability of over- and under-estimation. Concerning over-estimation, no boundedness or loss-of-memory conditions are required: the proof relies on new deviation inequalities for empirical probabilities of independent interest. The under-estimation properties rely on classical hypotheses for processes of infinite memory. These results improve on and generalize the bounds obtained in Duarte et al. (2006) [12], Galves et al. (2008) [18], Galves and Leonardi (2008) [17], Leonardi (2010) [22], refining asymptotic results of Bühlmann and Wyner (1999) [4] and Csiszár and Talata (2006) [9].     

Variable annuities: A unifying valuation approach

Life annuities and pension products usually involve a number of guarantees, such as minimum accumulation rates, minimum annual payments or a minimum total payout. Packaging different types of guarantees is the feature of so-called variable annuities. Basically, these products are unit-linked investment policies providing a post-retirement income. The guarantees, commonly referred to as GMxBs (namely, Guaranteed Minimum Benefits of type ‘x’), include minimum benefits both in the case of death and survival. In this paper we propose a unifying framework for the valuation of variable annuities under quite general model assumptions. We compute and compare contract values and fair fee rates under ‘static’ and ‘mixed’ valuation approaches, via ordinary and least squares Monte Carlo methods, respectively.     

Variable annuities: A unifying valuation approach

Life annuities and pension products usually involve a number of guarantees, such as minimum accumulation rates, minimum annual payments or a minimum total payout. Packaging different types of guarantees is the feature of so-called variable annuities. Basically, these products are unit-linked investment policies providing a post-retirement income. The guarantees, commonly referred to as GMxBs (namely, Guaranteed Minimum Benefits of type ‘x’), include minimum benefits both in the case of death and survival. In this paper we propose a unifying framework for the valuation of variable annuities under quite general model assumptions. We compute and compare contract values and fair fee rates under ‘static’ and ‘mixed’ valuation approaches, via ordinary and least squares Monte Carlo methods, respectively.