Time-varying Weights Research Articles

Study of the cross-market effects of Brexit based on the improved symbolic transfer entropy GARCH model-An empirical analysis of stock-bond correlations.

In this paper, we study the cross-market effects of Brexit on the stock and bond markets of nine major countries in the world. By incorporating information theory, we introduce the time-varying impact weights based on symbolic transfer entropy to improve the traditional GARCH model. The empirical results show that under the influence of Brexit, flight-to-quality not only commonly occurs between the stocks and bonds of each country but also simultaneously occurs among different countries. We also find that the accuracy of the time-varying symbolic transfer entropy GARCH model proposed in this paper has been improved compared to the traditional GARCH model, which indicates that it has a certain practical application value.

Pinning synchronization of fractional order complex-variable dynamical networks with time-varying coupling

This paper investigates the synchronization of fractional order complex-variable dynamical networks with time-varying coupling. Based on information of the complex network's configuration, an effective adaptive pinning control strategy to adjust simultaneously coupling strength and feedback gain is designed. Besides, we also consider the synchronization in complex networks with time-varying coupling weight. By constructing suitable Lyapunov function and using the presented lemma, some sufficient criteria are derived to achieve the synchronization of fractional order complex-variable dynamical networks under the corresponding update law. The update law is only dependent on the states of the complex dynamical networks, which do not need any other information such as the characteristic of the uncoupled nodes of the networks. Further, the result extends the synchronization condition of the real-variable dynamical networks to the complex-valued field. Finally, the correctness and feasibility of the proposed theoretical results are verified by two examples of fractional complex-variable dynamic networks.

Stability analysis for coupled systems with time-varying coupling structure

The stability of coupled systems with time-varying coupling structure (CSTCS) is considered in this paper. The graph-theoretic method on a digraph with constant weight has been successfully generalized into a digraph with time-varying weight. In addition, we construct a global Lyapunov function for CSTCS. By using the graph theory and the Lyapunov method, a Lyapunov-type theorem and some sufficient criteria are obtained. Furthermore, the theoretical conclusions on CSTCS can successfully be applied to the predator-prey model with time-varying dispersal. Finally, a numerical example of CSTCS is given to illustrate the effectiveness and feasibility of our results.

Online weight design for distributed filtering with limited power

In this study, the authors consider distributed estimation over a sensor network with limited power. Based on PageRank algorithm, they propose a distributed estimator, where the link weight is designed using online information. They compared the network performance of the proposed estimator with time-varying weight and constant weight under identical initial conditions. To extend the lifetime of sensors, they introduce an online power scheduling method to distributed estimation problem, where each sensor allocates the power for the communication link based on real-time estimation error. They also provide a sufficient condition to guarantee the stability of the networked system with power scheduling.

Trend inflation estimates for Thailand from disaggregated data

This paper constructs a new trend inflation measure for Thailand based on the multivariate unobserved components model with stochastic volatility and outlier adjustments (MUCSVO) of Stock and Watson (2016). Similar to core inflation, the MUCSVO produces an estimate of trend inflation utilizing information in disaggregated data, but also allows for time-varying weights that depend on the volatility, persistence and comovement of the underlying sectoral inflation series. Based on the empirical results, the majority of sectoral weights show significant time-variation in contrast to their relatively stable expenditure shares. Volatile food and energy sectors that are typically excluded from core inflation measures also turn out to help explain approximately 10 percent of MUCSVO trend inflation rate movements. Compared against other benchmark trend inflation measures, we show that the MUCSVO delivers trend estimates that are smoother, more precise, and are able to forecast average inflation over the 1–3 year horizon more accurately both in-sample and out-of-sample, especially since the year 2000.

Event-triggered output feedback synchronization control of complex dynamical networks

Abstract In this paper, the problem of event-triggered output feedback synchronization control for complex dynamical networks (CDN) is studied. Based on a distributed state observer, a novel event-triggered mechanism is firstly introduced to design the output feedback synchronization controller, and by applying a useful graph theoretical result, the asymptotic convergence of the synchronization errors is then guaranteed. Moreover, it is proved that the designed event-triggered synchronization method is also valid for the CDNs with time-varying coupling weights. In particular, the proposed method is an extension of the existing state feedback event-triggered synchronization results. Finally, the circuit realizations for the new event-triggered mechanism and control approach are given, and the effectiveness of the proposed scheme is illustrated through the simulation results.

A Finite-Time Consensus Framework Over Time-Varying Graph Topologies With Temporal Constraints

Finite-time consensus has attracted significant research interest due to its wide applications in multiagent systems. Various results have been developed to enable multiagent systems to complete desired tasks in finite-time. However, most existing results in the literature can only ensure finite-time consensus without considering temporal constraints, where the time used to achieve consensus cannot be preset arbitrarily and is generally determined by the system initial conditions, prohibiting its application in time-sensitive tasks. Motivated to achieve consensus within a desired time frame, user-specified finite-time consensus is developed in the present work for a multiagent system to ensure consensus at a prespecified time instant. The interaction among agents (e.g., communication and information exchange) is modeled as a time-varying graph, where each edge is associated with a time-varying weight representing the time-varying interaction between neighboring agents. Consensus over such time-varying graph is then proven based on a time transformation and is guaranteed to be completed within a prespecified time frame. To demonstrate the developed framework, finite-time rendezvous of a multiagent system is considered as an example application, where agents with limited communication capabilities are desired to meet at a common location at a preset time instant with constraints on preserving global network connectivity. A numerical simulation is provided to demonstrate the efficiency of the developed result.

Robust pinning control of heterogeneous complex networks with jointly connected topologies and time-varying parametric uncertainty

ABSTRACTThe pinning/leader control problems provide the design of the leader or pinning controller in order to guide a complex network to a desired trajectory or target (synchronisation or consensus). Let a time-invariant complex network, pinning/leader control problems include the design of the leader or pinning controller gain and number of nodes to pin in order to guide a network to a desired trajectory (synchronization or consensus). Usually, lower is the number of pinned nodes larger is the pinning gain required to assess network synchronisation. On the other side, realistic application scenario of complex networks is characterised by switching topologies, time-varying node coupling strength and link weight that make hard to solve the pinning/leader control problem. Additionally, the system dynamics at nodes can be heterogeneous. In this paper, we derive robust stabilisation conditions of time-varying heterogeneous complex networks with jointly connected topologies when coupling strength and link weight interactions are affected by time-varying uncertainties. By employing Lyapunov stability theory and linear matrix inequality (LMI) technique, we formulate low computationally demanding stabilisability conditions to design a pinning/leader control gain for robust network synchronisation. The effectiveness of the proposed approach is shown by several design examples applied to a paradigmatic well-known complex network composed of heterogeneous Chua's circuits.

Combined Density Nowcasting in an Uncertain Economic Environment

We introduce a combined density nowcasting (CDN) approach to dynamic factor models (DFM) that in a coherent way accounts for time-varying uncertainty of several model and data features to provide more accurate and complete density nowcasts. The combination weights are latent random variables that depend on past nowcasting performance and other learning mechanisms. The combined density scheme is incorporated in a Bayesian sequential Monte Carlo method which rebalances the set of nowcasted densities in each period using updated information on the time-varying weights. Experiments with simulated data show that CDN works particularly well in a situation of early data releases with relatively large data uncertainty and model incompleteness. Empirical results, based on U.S. real-time data of 120 monthly variables, indicate that CDN gives more accurate density nowcasts of U.S. GDP growth than a model selection strategy and other combination strategies throughout the quarter with relatively large gains for the two first months of the quarter. CDN also provides informative signals on model incompleteness during recent recessions. Focusing on the tails, CDN delivers probabilities of negative growth, that provide good signals for calling recessions and ending economic slumps in real time.

Sampling algorithms for stochastic graphs: A learning automata approach

Recently, there has been growing interest in social network analysis. Graph models for social network analysis are usually assumed to be a deterministic graph with fixed weights for its edges or nodes. As activities of users in online social networks are changed with time, however, this assumption is too restrictive because of uncertainty, unpredictability and the time-varying nature of such real networks. The existing network measures and network sampling algorithms for complex social networks are designed basically for deterministic binary graphs with fixed weights. This results in loss of much of the information about the behavior of the network contained in its time-varying edge weights of network, such that is not an appropriate measure or sample for unveiling the important natural properties of the original network embedded in the varying edge weights. In this paper, we suggest that using stochastic graphs, in which weights associated with the edges are random variables, can be a suitable model for complex social network. Once the network model is chosen to be stochastic graphs, every aspect of the network such as path, clique, spanning tree, network measures and sampling algorithms should be treated stochastically. In particular, the network measures should be reformulated and new network sampling algorithms must be designed to reflect the stochastic nature of the network. In this paper, we first define some network measures for stochastic graphs, and then we propose four sampling algorithms based on learning automata for stochastic graphs. In order to study the performance of the proposed sampling algorithms, several experiments are conducted on real and synthetic stochastic graphs. The performances of these algorithms are studied in terms of Kolmogorov-Smirnov D statistics, relative error, Kendall's rank correlation coefficient and relative cost.

Time-Varying Weighted Optimal Empirical Mode Decomposition Using Multiple Sets of Basis Functions

Empirical mode decomposition (EMD) is a favorite tool for analyzing nonlinear and non-stationary signals. It decomposes any signal into a finite set of oscillation modes consisting of intrinsic mode functions and a residual function. Superimposing all these modes reconstructs the signal without any information loss. In addition to satisfying the perfect reconstruction property, however, there is no implication about the reconstruction optimality of the EMD. The lack of optimality restricts the signal recovery capability of the EMD in the presence of disturbances. Only a few attempts are made to meet this deficiency. In this paper, we propose a new algorithm named as time-varying weighted EMD. By this algorithm, original signal is reconstructed in the minimum mean-square error sense through the EMD followed by time-varying weightings of the oscillation modes. Determining the time-varying weights for the oscillation modes constitutes the backbone of the algorithm. Aiming to determine the time-varying weights of the oscillation modes; we use multiple sets of basis functions. The effectiveness of the proposed algorithm is demonstrated by computer simulations involving real biomedical signals. Simulation results show that the proposed algorithm exhibits better performance than that of its existing counterparts in terms of lower mean-square error and higher signal-to-error ratio.

An investigation on the relationship between return and trading volume: asymmetric V-type or asymmetric increasing-type pattern

This paper develops a new mechanism that takes into account the fast change in behaviours of futures returns and trading volumes in order to model the time-varying and quantile-varying dependence between return and volume for energy-related futures products traded on TOCOM, NYMEX and ICE Futures Europe. A logistic function with the product of one-step-ahead expectations of return and volume as a transition variable is used to depict the time-varying weight of a mixture copula. This paper then employs a mixture copula of a Gumbel copula and a rotated Gumbel copula to detect the asymmetric V-type pattern and uses a mixture copula of a Gumbel copula and a survival Gumbel copula to measure the asymmetric increasing-type pattern. Empirical results demonstrate that the asymmetric V-type pattern is a more appropriate specification to characterize the return–volume nexus than the asymmetric increasing-type pattern, irrespective of the types of energy-related futures products and futures exchanges. The time-varying dependence has greater dependence in the lower–upper corner of the joint distribution than in the upper–upper corner of the joint distribution, implying that market participants believe that market reversals are more likely during periods of price declines than in periods of price increases. Moreover, this paper shows the inappropriateness of the two-step estimation method that has been widely used in the existing literature.

A Distributed Networked Approach for Fault Detection of Large-Scale Systems

Networked systems present some key new challenges in the development of fault-diagnosis architectures. This paper proposes a novel distributed networked fault detection methodology for large-scale interconnected systems. The proposed formulation incorporates a synchronization methodology with a filtering approach in order to reduce the effect of measurement noise and time delays on the fault detection performance. The proposed approach allows the monitoring of multirate systems, where asynchronous and delayed measurements are available. This is achieved through the development of a virtual sensor scheme with a model-based resynchronization algorithm and a delay compensation strategy for distributed fault-diagnostic units. The monitoring architecture exploits an adaptive approximator with learning capabilities for handling uncertainties in the interconnection dynamics. A consensus-based estimator with time-varying weights is introduced, for improving fault detectability in the case of variables shared among more than one subsystem. Furthermore, time-varying threshold functions are designed to prevent false-positive alarms. Analytical fault detectability sufficient conditions are derived, and extensive simulation results are presented to illustrate the effectiveness of the distributed fault detection technique.

How is the likelihood of fire sales in a crisis affected by the interaction of various bank regulations?

We present a model that describes how different types of bank regulation can interact to affect the likelihood of fire sales in a crisis. In our model, risk shifting motives drive how banks recapitalize following a negative shock, leading banks to concentrate their portfolios. Regulation affects the likelihood of fire sales by giving banks the incentive to sell certain assets and retain others. Ex-post incentives from high risk weights and the interaction of capital and liquidity requirements can make fire sales more likely. Time-varying risk weights may be an effective tool to prevent fire sales.

How is the Likelihood of Fire Sales in a Crisis Affected by the Interaction of Various Bank Regulations?

We present a model that describes how different types of bank regulation can interact to affect the likelihood of fire sales in a crisis. In our model, risk shifting motives drive how banks recapitalize following a negative shock, leading banks to concentrate their portfolios. Regulation affects the likelihood of fire sales by giving banks the incentive to sell certain assets and retain others. Ex-post incentives from high risk weights and the interaction of capital and liquidity requirements can make fire sales more likely. Time-varying risk weights may be an effective tool to prevent fire sales.

Expansion of the Decoupled Discreet-Time Jacobian Eigenvalue Approximation for Model-Free Analysis of PMU Data

This paper proposes an extension of the algorithm in [1], as well as utilization of the wavelet transform in event detection, including High Impedance Fault (HIF). Techniques to analyze the abundant data of PMUs quickly and effectively are paramount to increasing response time to events and unstable parameters. With the amount of data PMUs output, unstable parameters, tie line oscillations, and HIFs are often overlooked in the bulk of the data. This paper explores model-free techniques to attain stability information and determine events in real-time. When full system connectivity is unknown, many traditional methods requiring other bus measurements can be impossible or computationally extensive to apply. The traditional method of interest is analyzing the power flow Jacobian for singularities and system weak points, attained by applying singular value decomposition. This paper further develops upon the approach in [1] to expand the Discrete-Time Jacobian Eigenvalue Approximation (DDJEA), giving values to significant off-diagonal terms while establishing a generalized connectivity between correlated buses. Statistical linear models are applied over large data sets to prove significance to each term. Then the off diagonal terms are given time-varying weights to account for changes in topology or sensitivity to events using a reduced system model. The results of this novel method are compared to the present errors of the previous publication in order to quantify the degree of improvement that this novel method imposes. The effective bus eigenvalues are briefly compared to Prony analysis to check similarities. An additional application for biorthogonal wavelets is also introduced to detect event types, including the HIF, for PMU data.

Algorithm of key data ensemble clustering and approximate analysis in cloud computing

One collaborative data fusion recommendation algorithm is SFS-TOPSIS based on customer satisfaction degree. First, it starts from calculation efficiency and recommended precision angle that improves service recommendation algorithm, makes real-time updating and algorithm improvement for it by combining time-varying weight method TOPSIS fusion algorithm and designs a collaborative data fusion recommendation algorithm based on customer satisfaction degree. Second, for the problem of inadequate definition of traditional similarity for resolution, improvements have been made based on user evaluation confidence. Last, time-varying weight method has been adopted to improve standard TOPSIS fusion, improve time-varying attribute of TOPSIS decision fusion and realise effective attribute fusion of user similarity data; through making simulation comparison on two standard testing sets of MovieLens and BookCrossing, it indicates that the service recommendation performance of SFS-TOPSIS is superior. The proposed SFS-TOPSIS algorithm can improve service recommendation accuracy effectively and it is with certain application value.

Decision-making for Civil Infrastructures Incorporating the time-Varying Effect of Risk Preference

Life cycle cost (LCC) analysis offered a useful framework to think about the efficiency of infrastructures investment. However, the widely accepted algorithm of the LCC doesn’t account for the decision maker's attitude toward the risk of future consequences. The cumulative prospect theory (CPT) can be used to reveal the decision maker's preference under different risk and uncertainty conditions, and has been applied in literature to some design optimization issues. Nevertheless, previous research didn’t consider the fact that people's risk preference may vary with time, i.e., there is a reasonable tendency that people will be more risk-averse to the near future losses than to the far future losses. Recognizing this, this paper adopts the CPT model in the minimum expected cost analysis, and moreover incorporates the time-varying weighting function included in the CPT model to reflect the possible changing of risk preference toward future events. A seismic retrofit optimization of a dam built in strong earthquake-prone area is used to demonstrate the application of the method, and it is found that considering the time-varying of risk preference leads to a low design level compared with that by the CPT model, but a high design level compared with the traditional LCC.

State-Dependent Transmission of Monetary Policy in the Euro Area

In this paper, we estimate a logit mixture vector autoregressive (Logit-MVAR) model describing monetary policy transmission in the euro area over the period 1999-2015. MVARs allow us to differentiate between different states of the economy. In our model, the time-varying state weights are determined by an underlying logit model. In contrast to other classes of non-linear VARs, the regime affiliation is neither strictly binary, nor binary with a transition period, and based on multiple variables. We show that monetary policy transmission in the euro area can indeed be described as a mixture of two states. The first (second) state with an overall share of 84% (16%) can be interpreted as a “normal state” (“crisis state”). In both states, output and prices are found to decrease after monetary policy shocks. During “crisis times” the contraction is much stronger, as the peak effect is roughly one-and-a-half times as large when compared to “normal times.” In contrast, the effect of monetary policy shocks is less enduring in crisis times. Both findings provide a strong indication that the transmission mechanism is indeed different for the euro area during times of economic and financial distress.

Forecasting the realized range-based volatility using dynamic model averaging approach

In this study, we forecast the realized range-based volatility (RRV) using the heterogeneous autoregressive realized range-based volatility (HAR-RRV) model and its various extensions, which are called HAR-RRV-type models. We first consider the time-varying property of those models’ parameters using the dynamic model averaging (DMA) approach and evaluate the forecasting performance of three types: individual HAR-RRV-type models, combined models with constant weights, and combined models with time-varying weights. Our out-of-sample empirical results show that combined models with time-varying weights can not only generate more accurate forecasts, but also beat individual models and combined models with constant weights.