Stress testing network reconstruction via graphical causal model

An resilience optimal evaluation of financial portfolios implies having plausible hypotheses about the multiple interconnections between the macroeconomic variables and the risk parameters. In this article, we propose a graphical model for the reconstruction of the causal structure that links the multiple macroeconomic variables and the assessed risk parameters, it is this structure that we call stress testing network. In this model, the relationships between the macroeconomic variables and the risk parameter define a “relational graph” among their time‐series, where related time‐series are connected by an edge. Our proposal is based on the temporal causal models, but unlike, we incorporate specific conditions in the structure which correspond to intrinsic characteristics this type of networks. Using the proposed model and given the high‐dimensional nature of the problem, we used regularization methods to efficiently detect causality in the time‐series and reconstruct the underlying causal structure. In addition, we illustrate the use of model in credit risk data of a portfolio. Finally, we discuss its uses and practical benefits in stress testing.     

Identifiability of intermediate variables on causal paths

We discuss the discovery of causal mechanisms and identifiability of intermediate variables on a causal path. Different from variable selection, we try to distinguish intermediate variables on the causal path from other variables. It is also different from ordinary model selection approaches which do not concern the causal relationships and do not contain unobserved variables. We propose an approach for selecting a causal mechanism depicted by a directed acyclic graph (DAG) with an unobserved variable. We consider several causal networks, and discuss their identifiability by observed data. We show that causal mechanisms of linear structural equation models are not identifiable. Furthermore, we present that causal mechanisms of nonlinear models are identifiable, and we demonstrate the identifiability of causal mechanisms of quadratic equation models. Sensitivity analysis is conducted for the identifiability.     

Reasoning about the effects of actions in automated planning systems

This paper describes a temporal reasoning system that supports deductions for modeling the physics (i.e. cause and effect relationships) of a specified planning domain. We demonstrate how the process of planning can be profitably partitioned into two inferential components: one responsible for making choices relevant to the construction of a plan and a second responsible for maintaining an accurate picture of the future that takes into account the planner's intended actions. Causal knowledge about the effects of actions and the behavior of processes is stored apart from knowledge of plans for achieving specific tasks. Using this causal knowledge, the second component is able to predict the consequences of actions proposed by the first component and notice interactions that may affect the success of the plan under construction. By keeping track of the reasons why each prediction and choice is made, the resulting system is able to reason efficiently about the consequences of making new choices and retracting old ones. The system described in this paper makes it particularly simple and efficient to reason about actions whose effects vary depending upon the circumstances in which the actions are executed.     

Aligning General Chemistry Laboratory With Lecture at a Large University

The laboratory and lecture components of general chemistry are commonly offered as two separate courses, with lecture typically meeting two or three times per week and laboratory scheduled to meet only once per week. The concepts, content, and relationships presented in lecture may be disjointed and asynchronous with respect to those encountered in laboratory experiments. In addition, traditional laboratory experiments tend to be confirmation labs, in which students are aware of the “right” answer before beginning the lab. Students enrolled in a specific lecture section do not necessarily meet for the same laboratory section. As such, learning experiences in laboratory do little to help the students construct an understanding of chemical concepts, content, or relationships. The goal of this project was to develop an inquiry‐based approach to curriculum and instruction in first‐semester general chemistry at the University of North Carolina at Charlotte. A major objective of the project was to develop a laboratory curriculum that meshes intimately with lecture. This objective was accomplished by (a) creating a laboratory course that met for 80‐minutes twice a week immediately following the lecture, (b) involving students in laboratory experiments that related to the material presented during lecture, and (c) using laboratory observations and data in lecture to help students construct an understanding of chemical phenomena.     

Identifying the Causal Effects of Stochastic Interventions by Augmented Causal Networks

Suppose that the cause-effect relationships between variables can be described by a causal network. To identify the causal effect of a stochastic intervention, an augmented causal network for stochastic intervention is proposed in this paper. Then, we obtain two graphical criteria for identifying the causal effects of stochastic interventions from passive observations on observed variables only. When either of the two criteria is satisfied, a simple closed-form expression is provided for the causal effect of a stochastic intervention, which enables researchers to assess the causal effect with little effort.     

Quadratic integral games and causal synthesis

The game problem for an input-output system governed by a Volterra integral equation with respect to a quadratic performance functional is an untouched open problem. In this paper, it is studied by a new approach called projection causality. The main result is the causal synthesis which provides a causal feedback implementation of the optimal strategies in the saddle point sense. The linear feedback operator is determined by the solution of a Fredholm integral operator equation, which is independent of data functions and control functions. Two application examples are included. The first one is quadratic differential games of a linear system with arbitrary finite delays in the state variable and control variables. The second is the standard linear-quadratic differential games, for which it is proved that the causal synthesis can be reduced to a known result where the feedback operator is determined by the solution of a differential Riccati operator equation.

     

Semiotic open complex systems: Processes and behaviors

An H‐system is a conceptual or semiotic model of reality formed in the mind of the subject. Understanding the behavior of the system means being able to infer causal relationships that explain this system to the Observer‐subject, and therefore having access to mechanisms to construct a mental or ontological mathematical model of the system under study. A process is a mechanism involving a series of successive operations between stimuli and responses. © 2016 Wiley Periodicals, Inc. Complexity 21: 388–396, 2016     

From vicious to virtuous circles: problem structuring for quantified decision making in operationalization of corporate social responsibility

There is no formalised approach for problem structuring and quantitative decision support to operationalise corporate social responsibility (CSR) implementation. In this paper, techniques for considering criteria relationships are outlined and a holistic, systematic framework combining a qualitative and quantitative method for practical CSR integration is provided. Cognitive mapping (CM) is applied to structure the problem picture, and the cause–effect relationships between decision elements. Soft CM methodology is employed to assess the cross-criteria interactions, at both an individual and a collective level. The interactions of criteria can have a significant impact upon CSR implementation. Such impacts can be direct or indirect through their close linkages to other criteria. The causal strategic map serves as an input to the analytic network process (ANP) to carry out the multi-criteria decision analysis. Then, CM and ANP are applied in a comparative analysis to verify whether the measures of criteria significance do correspond. The key criteria in networks are identified using centrality in CM and single limited priorities in ANP. This study demonstrates that using criteria without considering their interactions will result in shortcomings in the evaluation and assessment of CSR programmes. The holistic framework, combining CM and ANP proposed in this work, enhances the process of problem structuring and supports preference-based evaluation of decision alternatives. The results of our study yield that the mapping procedure has an influence on the criteria significance in networks. The correspondence between CM and ANP is stronger when cause relationships are rigidly interpreted. More unambiguous interpretations of causal relations can be achieved if methods are used jointly and common peaks of importance in both CM and ANP could potentially serve as indications of key decision elements.     

A METHOD FOR SIMPLIFYING LARGE ECOSYSTEM MODELS

Abstract Simplifying large ecosystem models is essential if we are to understand the underlying causes of observed behaviors. However, such understanding is often employed to achieve simplification. This paper introduces two model simplification methods that can be applied without requiring intimate prior knowledge of the system. Their utility is measured by the resulting values of given model diagnostics relative to those of the large model. The first method is a simple automated procedure for nondimensionalizing large ecosystem models, which identifies and eliminates terms that have little effect on model diagnostics. Some of its limitations are then addressed by the rate elimination method, which measures the relative importance of model terms using least‐squares regression. The methods are applied to a model of the nitrogen cycle in Port Phillip Bay, Victoria, Australia. The rate elimination method provided more insights into the causal relationships built into the model than the nondimensionalizing method. It also allowed the reduction of the model's dimension. Thus it is a useful first step in model simplification.     

Linking structural equation modeling to Bayesian networks: Decision support for customer retention in virtual communities

Bayesian networks are limited in differentiating between causal and spurious relationships among decision factors. Decision making without differentiating the two relationships cannot be effective. To overcome this limitation of Bayesian networks, this study proposes linking Bayesian networks to structural equation modeling (SEM), which has an advantage in testing causal relationships between factors. The capability of SEM in empirical validation combined with the prediction and diagnosis capabilities of Bayesian modeling facilitates effective decision making from identification of causal relationships to decision support. This study applies the proposed integrated approach to decision support for customer retention in a virtual community. The application results provide insights for practitioners on how to retain their customers. This research benefits Bayesian researchers by providing the application of modeling causal relationships at latent variable level, and helps SEM researchers in extending their models for managerial prediction and diagnosis.     

Direct realism: Proximate causation and the missing object

Direct Realists believe that perception involves direct awareness of an object not dependent for its existence on the perceiver. Howard Robinson rejects this doctrine in favour of a Sense-Datum theory of perception. His argument against Direct Realism invokes the principle ‘same proximate cause, same immediate effect’. Since there are cases in which direct awareness has the same proximate cerebral cause as awareness of a sense datum, the Direct Realist is, he thinks, obliged to deny this causal principle. I suggest that although Direct Realism is in more than one respect implausible, it does not succumb to Robinson’s argument. The causal principle is true only if ‘proximate cause’ means ‘proximate sufficient cause’, and the Direct Realist need not concede that there is a sufficient cerebral cause for direct awareness of independent objects.     

An inverse problem of identifying the diffusion coefficient in a coupled parabolic‐elliptic system

This work studies a nonlinear inverse problem of reconstructing the diffusion coefficient in a parabolic‐elliptic system using the final measurement data, which has important application in a large field of applied science. Being different from other works, which are governed by single partial differential equations, the underlying mathematical model in this paper is a coupled parabolic‐elliptic system, which makes theoretical analysis rather difficult. On the basis of the optimal control framework, the identification problem is transformed into an optimization problem. Then the existence of the minimizer is proved, and the necessary condition that must be satisfied by the minimizer is also given. Since the optimal control problem is nonconvex, one may not expect a unique solution universally. However, the local uniqueness and stability of the minimizer are deduced in this paper.     

Inferring social structure from continuous‐time interaction data

Relational event data, which consist of events involving pairs of actors over time, are now commonly available at the finest of temporal resolutions. Existing continuous‐time methods for modeling such data are based on point processes and directly model interaction “contagion,” whereby one interaction increases the propensity of future interactions among actors, often as dictated by some latent variable structure. In this article, we present an alternative approach to using temporal‐relational point process models for continuous‐time event data. We characterize interactions between a pair of actors as either spurious or as resulting from an underlying, persistent connection in a latent social network. We argue that consistent deviations from expected behavior, rather than solely high frequency counts, are crucial for identifying well‐established underlying social relationships. This study aims to explore these latent network structures in two contexts: one comprising of college students and another involving barn swallows.     

中国人均碳排放影响因素的长期均衡与因果动态关系研究——基于结构突变ARDL-VECM模型的实证分析

为了探讨中国低碳经济的发展路径,运用BP内生结构突变点检验、考虑结构突变的ARDL模型和基于VECM模型的Granger因果关系检验方法,利用我国1985至2014年的相关数据,探讨经济增长、城镇化、技术创新、贸易开放与我国人均碳排放的长期均衡和短期动态关系,以及变量之间的因果关系。研究结果表明,中国人均碳排放与解释变量之间存在长期协整关系。长期来看,经济增长与人均碳排放之间呈倒U型关系,城市化水平的提高可以有效减少人均碳排放,技术创新与人均碳排放显著负相关,而贸易开放会引起环境恶化。短期来看,贸易开放与人均碳排放显著正相关,而其他变量对人均碳排放的影响均不显著。此外,基于VECM模型的Granger因果关系检验证实所有解释变量均为人均碳排放的Granger原因,且经济增长、贸易开放与人均碳排放之间存在双向因果关系。     

An Extension Principle for the Einstein-Vlasov System in Spherical Symmetry

We prove that “first singularities” in the non-trapped region of the maximal development of spherically symmetric asymptotically flat data for the Einstein-Vlasov system must necessarily emanate from the center. The notion of “first” depends only on the causal structure and can be described in the language of terminal indecomposable pasts (TIPs). This result suggests a local approach to proving weak cosmic censorship for this system. It can also be used to give the first proof of the formation of black holes by the collapse of collisionless matter from regular initial configurations. Communicated by Sergiu Klainerman submitted 15/11/04, accepted 17/02/05     

The Pathmox approach for PLS path modeling: Discovering which constructs differentiate segments

The problem of heterogeneity represents a very important issue in the decision‐making process. Furthermore, it has become common practice in the context of marketing research to assume that different population parameters are possible depending on sociodemographic and psycho‐demographic variables such as age, gender, and social status. In recent decades, numerous approaches have been proposed with the aim of involving heterogeneity in the parameter estimation procedures. In partial least squares path modeling, the common practice consists of achieving a global measurement of the differences arising from heterogeneity. This leaves the analyst with the important task of detecting, a posteriori, which are the causal relationships (ie, path coefficients) that produce changes in the model. This is the case in Pathmox analysis, which solves the heterogeneity problem by building a binary tree to detect those segments of population that cause the heterogeneity. In this article, we propose extending the same Pathmox methodology to asses which particular endogenous equation of the structural model and which path coefficients are responsible of the difference.     

Modeling Dependencies in Operational Risk with Hybrid Bayesian Networks

This paper addresses the problem of quantifying and modeling financial institutions’ operational risk in accordance with the Advanced Measurement Approach put forth in the Basel II Accord. We argue that standard approaches focusing on modeling stochastic dependencies are not sufficient to adequately assess operational risk. In addition to stochastic dependencies, causal topological dependencies between the risk classes are typically encountered. These dependencies arise when risk units have common information- and/or work-flows and when failure of upstream processes imply risk for downstream processes. In this paper, we present a modeling strategy that explicitly captures both topological and stochastic dependencies between risk classes. We represent the operational-risk taxonomy in the framework of a hybrid Bayesian network (BN) and provide an intuitively compelling approach for handling causal relationships and external influences. We demonstrate the use of hybrid BNs as a tool for mapping causal dependencies between frequencies and severities of risk events and for modeling common shocks. Monte-Carlo simulations illustrate that the impact of topological dependencies on triggering overall system breakdowns can be substantial.     

Generalization of the Maxwell equation and relation to the elastic equation

In this paper, we propose a hyperbolic system of first‐order pseudo‐differential equations as generalization of the Maxwell equation. We state basic properties of this system corresponding to the ones of the (usual) Maxwell equation and explain that several known generalized Maxwell equations presented by some researchers can be integrated into the system. Namely, their equations can be regarded as our equation in special cases. Their generalized equations admit not only transversal but also longitudinal waves and are examined from the physical viewpoint. Using the present system, from the mathematical viewpoint, we interpret the meaning for presence of the longitudinal wave (with the transversal one) in their generalized equations. This presence means existence of more than one non‐zero characteristic root for the system (ie, non‐zero eigenvalue of the symbol). We prove also that our system becomes a first‐order expression of (generalized) elastic equations. Furthermore, it is shown that introducing the elastic equations implies expressing the generalized Maxwell equations by the potentials.     

Learning dynamic causal relationships among sugar prices

In this paper, we are interested in exploring the dynamic causal relationships among two sets of three variables in different quarters. One set is futures sugar closing price in Zhengzhou futures exchange market (ZC), spot sugar price in Zhengzhou (ZS) and futures sugar closing price in New York futures exchange market(NC) and the other includes futures sugar opening price in Zhengzhou (ZO), ZS and NC. For each quarter, we first use Bayesian model selection to obtain the optimal causal graph with the highest BD scores and then use Bayesian model averaging approach to explore the causal relationship between every two variables. From the real data analysis, the two conclusions almost coincide, which shows that the two methods are practical.